From 606295d39f9f037d7f36a867c64c112c26ee0681 Mon Sep 17 00:00:00 2001
From: redoules <guillaume.redoules@gadz.org>
Date: Wed, 7 Nov 2018 21:12:06 +0100
Subject: [PATCH] added 3 new articles

bash functions
edges in a graph
reverse list
---
 .../Installing_bitcoind_on_raspberry_pi.html  |   2 +-
 cryptocurrencies/blockchain_bad.html          |   2 +-
 .../complete_graph.png                        | Bin 0 -> 104036 bytes
 index.html                                    |   6 +-
 jupyter/clear_cell.html                       |   2 +-
 jupyter/remote_run_notebook.html              |   2 +-
 linux/Reloading_.bashrc.html                  |   2 +-
 linux/Reloading_fstab.html                    |   2 +-
 linux/bashrc.html                             |   2 +-
 linux/directory_size.html                     |   2 +-
 linux/free_disk_space_linux.html              |   2 +-
 linux/get_ip_linux.html                       |   2 +-
 linux/linux_process_information.html          |   2 +-
 linux/mount_nfs_share_fstab.html              |   2 +-
 linux/share_nfs_share.html                    |   2 +-
 linux/simple_bash_function.html               | 257 ++++++++++++++++++
 linux/version_kernel.html                     |   2 +-
 ...he recommandation engine for articles.html |   2 +-
 mathematics/Number_edges_Complete_graph.html  | 190 +++++++++++++
 pages/about.html                              |   2 +-
 python/Creating_a_sqlite_database.html        |   2 +-
 ...of_eigenvalues_from_a_list_of_tensors.html |   2 +-
 python/Iterating_over_a_dataframe.html        |   2 +-
 python/Moving_average_pandas.html             |   2 +-
 python/Opening_SQLite_database.html           |   2 +-
 python/Opening_file.html                      |   2 +-
 python/Optimized_numpy_random_intel.html      |   2 +-
 ..._data_from_a_sql_database_with_pandas.html |   2 +-
 ...plotlib_figure_with_a_high_resolution.html |   2 +-
 ...notebook_for_plotting_with_matplotlib.html |   2 +-
 ...ng_data_to_a_sql_database_with_pandas.html |   2 +-
 python/dask_distributed_parallelism.html      |   2 +-
 python/dask_infiniband.html                   |   2 +-
 python/design_own_libs.html                   |   2 +-
 python/download_page.html                     |   2 +-
 python/ensure_dir.html                        |   2 +-
 python/list_files_directory.html              |   2 +-
 python/logplot.html                           |   2 +-
 python/reverse.html                           | 222 +++++++++++++++
 python/stock_pandas.html                      |   2 +-
 ...ting_all_python_package_with_anaconda.html |   2 +-
 search.html                                   |   2 +-
 sitemap.xml                                   |  23 +-
 sql/Sorting_results.html                      |   2 +-
 sql/WHERE_SQL_keywords.html                   |   2 +-
 sql/display_table.html                        |   2 +-
 sql/display_table_filter.html                 |   2 +-
 tipuesearch_content.json                      |   2 +-
 48 files changed, 737 insertions(+), 45 deletions(-)
 create mode 100644 images/Number_of_edges_in_a_Complete_graph/complete_graph.png
 create mode 100644 linux/simple_bash_function.html
 create mode 100644 mathematics/Number_edges_Complete_graph.html
 create mode 100644 python/reverse.html

diff --git a/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html b/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html
index 84a0770..fe778bb 100644
--- a/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html
+++ b/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html
@@ -286,7 +286,7 @@ maxuploadtarget=5000
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/cryptocurrencies/blockchain_bad.html b/cryptocurrencies/blockchain_bad.html
index a51e426..df7df7e 100644
--- a/cryptocurrencies/blockchain_bad.html
+++ b/cryptocurrencies/blockchain_bad.html
@@ -161,7 +161,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/images/Number_of_edges_in_a_Complete_graph/complete_graph.png b/images/Number_of_edges_in_a_Complete_graph/complete_graph.png
new file mode 100644
index 0000000000000000000000000000000000000000..06558f090ca4a91c78bb277c5b323f8e9b5311bd
GIT binary patch
literal 104036
zcmce-cT`hd^ENC>lirc8fC?f-fl#C>ML??3ODIZ9XwpkSK&69#^lB*5dx8iAkfze4
zhag3ePACGQgz_Hl=jW}@`~LH-udPKED<}KR*?acvnQLaQ6QQG}N=e2_cHzPWO0|bd
zdKWHSgaRK1QexoB`=726z&`?aJ=OacD*D-1fG>o0ikgZSE>y>nA6pRt-!Hp9G<Lsm
zfvWxdL(uI~^z_1ovjR0G#V0-%>sgmG89Mj6+t`Q*f70n_c{Xo-^09mpP2AMVXE7TQ
zY2QJvqZ@s5I-FLs>5j@B75WD<@Iia)Xv1g~I`JE<iC1-dpHaD=B#yMSG@CXenk)T3
z+I_V3v*pSe*l*V|9C)O?zh>Ctx2f;RoOiaHb2=P!q<yv`fAao}>_nR=<FD_#;I^}3
z%Rj#Zg9Q%c8KJ+&|8q0Z+4~U9|J?r{VU@FIPpbZr_ggsQaSk7P<D*M4oG%ji(|a~|
z+DV3_4myj6`zT^Hwj(epbMy$|&YQQ;Z|}du6&xu2zQ_nwf2V`j3xNN=NCqm<DT49K
z>HfY*l%dBrf?~+|_hN`9KZ7GUQv2VFZG{YE=feNJXerQX>Yc{VqmoLCO|{wx68Z=$
zF+NV-IxV~HKhm33)9gSd=f2w^WxL%YuyDf4fDqoY1NY4Mjmmf}#n_Hj)Ay$Gd#6df
zg0ZpyKP|UR`wf{5&$XyTXFs)_IxI03!j9(_!GcxsHBI;uW4|4C_g){U;L?DQ$Sw5H
z)<&qUZPX7<<6XG#`)BytD4a;%%>mJ<oIoC~_Y468yM?s2>#qi0R)QPdOIQXOuFT)H
znju!k_T<52o4-Y~agicgyYAmC2`P8i5yYD?c?^vqO;{imzFO#dMP2+ad$3>=@~4=n
zTvcu1oyUFR8Wv)e<!|;ON>8c^-gz?6Mzc>po1C;=X&CPP9yY!4!G%bs(GqIE>z?C0
z!6e!GrERTnLfb1{w*<pHd5lh3&?LSXkVA&-SJYpJG&G!j_qAQAS&VM|^=v=Q=F=v-
zRiS*9OpVJt9M=g7I*qzYJUHoA%Z2qokps`mH7BW`({~?^5bf1W(@S#6@Q>x-2<WnR
z?@f^8p_20ss11cMLUAhgAJH<SR%cyG*q#&^;$4WQGvrB$6PT;Q$l1c*K9+?!U}@QA
zQ?ybIFUIEx=ESZoO!l4z44{1%i?tZ^-F4dx5*C8*5UJ2z`TAjkysylyZ6^5R0MoX*
z&6|PLhATD-ByD3~p6?Ffx(ZBiqp!(84dr)=kub#J9b8V+>=&(PHjZG_I;6lC>h<_Q
z)~4{0=;N-fef!$8^~PR}bUI4l5L{aQdVfi@tyk7t#a_he-3+?}quniZK9*7gvEHLd
zl<|G+%}6WiQ^*z(Nygw0Zz7JvR|EEHs(E_ic)u>QZ+cu(qK~RuteCVNm|Qh)%Arjp
z#+QpM1=@avEsC-IK)+g?V2n#Qey6DJ$uf`%>6!@HcuqK$$ZAEdgO9AORIry78(5c@
zws*lyKWb?>p76ef{)trKje@}r#`0`8Iu;*x|5O$>xLpHn9t@}5smsq({PGlB-gIR*
zi3ML*zp-}Zn@@nnjyIn31p_G~G@?vQ!%l+j3z|q6=`TBn#w+QcjWpDy@Kn{krW`?~
zgT1RNp9$`<mA{q8Yn?C7=jk>$_$&=Q#B}$|_b1!yyLP>1#8s;;tZ#4QM!Rasaiq8}
z7*#TYm=qV$`F5u)fn3^sZKBD@n|);>)?*`%ygghw!*?Cua+n;3Xr^AraHBnv?L!mV
zP!=D3wR(OD_4L^+fhNC{h1D?NkpCEfWX(b4d*<|pIC{ovT#}oD0%3jxYraG#)+Mzw
zaWqK$_OsnME;;Xvv`1N{i*hyP2F_oz1U;P%s%WRys)B6bA%e+CiZ>^Qs$)_uD}oxH
z`9u1%zaPRlq-Bh&3T|T-YD3u+#;p=3S)+cf3;bO14L1Ma@B*4hm2EEe?$Ez`2JOI}
zrHn0o2`l$OG=S{qoE8hlu>9+5D^aG>tFpbl*ZWqEyg761x?<MMM2|*L3F~Y6?gM=T
zsRbNIpFK_?zLqw<GMsm|yS{<iPw&t~J=QyDQUMI*WumZm8s5j$2$}ToEu6cJxMxGO
z_Qq-=6>B=#Iys;YJ&~^-JS;2grQ$VA?>sv&mjMfay4Zo8L}gU9sX78mzp2}RT{60v
z9yyu6ZE)!26@}?8rC%d+1jl1zVa;_vkvdJ@>pa8zaho#UhLAI~m~BZK{>>T*8y6!~
z>am*olNDa7N;(4>hYI6wzP3{fF&M+k22EC+jjnp`VQ9J)!=XGJV|&kjT1K`rm7G17
zvD_{6bCsCtyTBrS+P}B_OdGSFexEFZUg5ry1ji-4PWF4}u3z;`8LOcZyWyQZ(eC>$
zM9_|t-0q8{6zn>^BiOBsg7{Kzg`>MqpJ;79EJ<<%#haXgw*#dOc2C=8Q?7B(h)8Ho
zw3~gSqtykTuFp#8_%@CV`Q6p<z`0O#lz8D4Cgss&x`C5Np`F+w@TC``VOuwnBc67L
z-Y#{G%Wy`Y$`M=uI09g<|Bpk6Ub5w524!8~uSDZ}i%}bHAn3$Yw<-5`3_6ci|L+H;
zq1w6n&9H|~k<tGh<C1#0P7iL1LdJXxy*DWz#c{Sj+TVE4-eNb9&5Tsr)WzuhcYvGo
zb?J2J3y(KVbDq)ynSU%JB~?=S9BDbJX8ZGVx%iyH#s8kWK(@Kp#AF%jNIzBwGO$_-
zcJ`X`1aCL1i~et_?5&me>EB&%(7)PyI<8oR*q*V>5$tmxXt^5r8V*F=EB^*q*q4Bf
zkf^?rQVZ_zI$RkNqr3=@5@wWtpbJcu;!iAo8Y};!PT|eem}0KyPXqT)yTcs=I!~(j
zi5~op{|MnU4WZ^<K*SPZ@L2Y!VUFD**z45+OC+Eq%YS1&?)pxip@q1tACxY<=4|ci
z5ayu8GAE`_pz_q{m_ixA2qOQ10|iSGJu~F-7zP$Zr4MduU!-=-#GUZnSg#4B?=F5*
zUOAw0=r8@>>YI-Xnxx=tiZHgzx_0fsYq2z!ApB)FaDl8N_-@eQ+Dz}u%l;YhJoR^`
z8o3xq>$vV&V@E9~gKbBOoIb(OhDU!t{QA!PxP|Ci#XXs2rIvY*BJG!x7N*EY0dYgG
zF@~x%mh?BZ?y7RY>JE3nr;bc!{>?0M)~2<K>1(f}uQZe!e{hWl9w-O8`6$P^D=vp-
z<0j0&PW2D|xxVv7d&BOd*U3kovGf(p9>lhOikfC7ukIagl+&>6<b>-CI_`C{7j^%;
z)7imN%agk6xqrY?!pd|r`0BUQ?}}B`F`yM$bv8$^6K{n0hl2!-`|wHqy3V}61{(*#
zY=1S=(?Yhi&^MY@9Ufpy)AXz<=Lp$~>=Vy-`F{}CmrUP9nOYpC_8ut+*_Xa2wf&Fa
zl@3<k%OIV>G1@=2m_)%bzeSs;IbMM$x{r;2Q(?Y;hey@9Jj<3$x)D~u2+>im3XVkL
ziYy<K0UiJVjFm&*q_i@9awk0N)vg`)fgFHs$(lgbgo{7qK`SVmvi&h&fwVAAKeU**
zljA9Fd&abWTVm4QeP3UC$QUb7Y7|-*Liwal{I3PhS#gwUf*kHr%O5`s10e^y<OV?n
zG5z-7g>doD+<@?mKlT@C*M=BRhv6m+p_eddRmaEaD^Q$%g;8i(2&HtR{o_B{xVu??
zP`)zL(sH=6RvCp=V8d5cj#PLE;Y*at6s3UCwf{AGSX*W=Gw_JTI)nUtV4(*mzT`=n
zVgoR+^zVTjhd+Ac?}idn-sL*o*tc#3MhkDoWA5F{YD4{^f#^h7hW7JefOGZg{y*o+
zP}OS^=DpJ?Jk5(#`)8>1y%=Y(Z-jW!lPX0!dZ#~nE0qN|Y|7R4fhM1>227;vZ&?&K
ziw=LrCA~Kal_6yFxG`Ax$0{%f%L74o+Et@Pv;&v;t7R`wxBDMb6gSrn72%TVjY6A%
z8<_s<hTPgr8Q&UoGVyeVVb~;}7#wy;19|fvnD5oU=PSB-{A<cJA~xr2&Z%0KvMFU?
zwl*Kv|1fVNMYx*SmmFm8@weD?sb8OY`h1uA=qGDh?Y4~dOzraH*L&U@UE&I%WHqiU
z{$oKw8Fr|%G8So%qQm(dIl9T(^YH<NBhwT14ESGyW34Hcp(>33DkQ&ll9^+I>fXBL
z#_^E3wZ9ZF_(qN(!utkm8>-~mSSv7f6^6fuk7K=WgYIoMxmV0Z4$O%jLIM5hZsPO>
ztBFxWY+~tu_7Sk$z_T#<bV?Aq^?0^nsS~F@3yeoSaNN8-QVY|6T&5^a$R?KXUw8CN
z*;Pm#rP_2LWoBy+sfx>Mbt-ZCPSwPtbkJ{naLwP|!NUN5G%+wumNI)VS{cNHTq4Nn
z^WHMo^S8_8#FjEyQ<m_fb7_^%t@@wzYNZV4IU7upc9>#*1G#W{rp;T;DJ$5ow$cPJ
z5~O9QAM>9@iekpky<{{iLRVj!4;8vOcdr1a@;-aLG})Tc@ftce`=8VDAGuK6-uBFy
z6%ykpurrF^WS)ezdT(b1)Q?Pm9*KgDnv^M)6S9r`W!_nRC?>&=ZR-u=7S<Ws2U;jo
zN66FFHdN%*vE1Yk%2nNew==Q-DVVG5_*+@a>CvKUWz#Jr{^jP+Mk(UrW?lY|Fh*rS
zHWAxM#$O8jg0k(G8O;p%W!H&xZ1r}h&@;``dy2o6U=+$t3M#t!$Eu{GV9iKq8jGZB
zuJQKdekbmq;p|Z()8P!GsK^^*xqQH6l>eHHHP}OjoY~?gQlCCH*d`vs8qbX}r0ry7
zP^(gu4F~;pd_yejYi%w8?NHR=;-U-JnK`ub2K%M%Pj}R?F)D-WrA5}1kvGt}%YXJ#
z>?7=C&xI=^Mq;`{=4iRT1834&?fQs4V`Tav!xvQK?XleO5X#8g|DGVVyvcD!LkkwK
zAw5&UAE}a{;_`UDNEoAenW8Qc8-xCzNvj|$7dFd9E)i3RRJNSHMq6k|-iVE{d5d9t
z#f>(U4HxIDu?*#o{4@E_IC1QU`QG&j_YNd?YqeW6pg<i6zJ$4&SeY5BBlbsutfMH4
zfR!-Yt=YE4_L&n4sPIL0w(if>>ev_sL2PMu2kS^EAUooZ>>}LW7wy2I?|y4e)~4~Z
zSELMz%V~>}V51pEp?<Vb9frSz527}VYdWCyQ(m5fy6(~J0^PR}<M$9%m3|Bm9c@4x
zfBcDB-o&^nkgBpsLsNZ^8fWMyqe#2I-f>^o^Ot?>gfB6yCf)-SKK7Tw(_yfU5b|ED
zFpe8YMwZw#n-t7ynj^+gHcGrmwT&3@*NKl`U&{icAV+&^c7fZtd)EH(KT@T<0z*eE
zDxdVARsoealYxqo{@7s5Q}7;VLCN^CCEaLMr3&D#2SR{T^VjB~$M>@jQUCa;Vbvg}
z$u)x5teIAqc+48@3~rRqnsU{=N)hqrj5!eE;feNd(~`Yj3$&;?R!Rh540$5N?Frk6
zNB^2B?fTl-q_afhdX2_|aQ47UEw5&H5#wyA_0qu(R<TISP{%)ep|^?K!^<BJQm~4B
zZrYmDu=bbz<nlvedomyA_bCh}4mS!_p@ZsB{@Ib$Db!-GR1f<-r*yQzz*feAlb)$;
zd_7FO1Q<#Woi+VqbbiQ6iJ4>L4-8$u_6@?ZdlW4ZGhHrkF>9}$udk5|$hydRw{_Yc
zMZC{Ahk{MVd4`{`F@TZN60tpm$3VU!7FgZObNo+VQEh63{&rx+c6j&>YVNB*CsO6b
zKjKzj-NqpRh@Ei`cL5liWb8Sr@hC;8K6TUppSdz<Ltndz3QN08k8?JvCXTs<egWx1
z%(H5xy1Ke*W`sce6??={k{=fH`YIYX*f1}4PbHsJ`MyxTwm-S)I^4Mb<^^T%DPFs{
zr&AE3Xo4Sd|1MR$`*FPV%x}HAyRS%>Ev@AgYg64!J6pfac32Kr+nX`e=~lOV8ZA#f
zsI)Rr|69bF@YQ8k(?j`lRNDJFe$pq$(@2BWxQbvr7SgnsZbve{D<2Oz2G@46rjU&L
zKz6$2VZ12teInC-8VuN_`CV}2qRR2rr-;!Ciu&gIx%xT(x1b?(O7?ue8s~U3=gIht
z+-c)Pz)oLZ=*+=u^z_3mU@FJ9#m&q4h)Uzye$lCqFi@3MJ-v(e{su`J(RiRmKoHNj
z<q+&Dgel!~OiHN<ne04en_Am14Gw5|NrudnTF$iz(H)NA5b9kmow1~n-Ss&Kc@FCu
z18z&aacdTFr)x)R%xSXw7!4>y>EQfv<;(fi!j75YRNgB!@KC|}rEc4UaHoy|dqP(8
zDoQq#rz>TOa)ra&g+lEd%Fca_BFi2YJ@LN0v*lNZS3iibg#&5NLmRNH9|gt0dym*g
z&ozipmOR)&;RDiboFn56^tyWUVdV_!Eco1E`#MeB1@!)-{#V8!R?k0P%?3OoxVL=g
zQPq5GS;OK-+d(*62mamk#M*EURXk>uqhi7YvPW0x@Rn2dbs^FdNSj0XqSl@~sY<#L
z1KV3~2=tLFa4!ofcUy@L!Pe%;UROGgNqgQ)g6QR4{i;!}Tj<r7q14$-h4ye}<zdei
zV6J|#_ip-{?fd-#SDqepTjj5WCUHe#V<=kEX8NVxV4~Dn@s1i1sPVSs$(<6GJI_tG
zMOlOU<LO|6&o%RIQ$t(}{-cJ-oDA~3N>U->!3$)0!V)J(l;a@>41hXiyR#ByD*R;+
zEBz`KzJ~ghO=&>DwRQu*mS!=aUF$Bn<oWMk>ihI6#{Y!Pyvn?G_Twz87E@(;foc89
z;(Dzx&5ZE7kiJVUEjt>bZgV5949cu;TV@Uy8s|{z{zI65jHqC|+2VM79Chkh+sR~_
z8h3U|UEM6&>axV{DwL1oP8dTOjfGz4FaV$Y%AG%T$>*^%jx#<_hT1#b3-%>0#t5O)
zQ})oe%;2vem{Zo;vhL$qR+qdUKfb*v0MjS>RSC1puEooq9@KZ@`U#g_xVK5gW1>`;
z@bKiA&d2_^??dP5q~gs$jjOYZX`Heks&VF#SMF=f-apKk6c^K<J7qBUH}id1G)<lA
z*6fw3R{0HqMnYlcft`$tFD}XPAU#zinKS8&%kAIVfCpN(;0;@FjZfW7=3a+Mi%EtV
zYdE8asW;`>eMfYU>`SCWSUcynC9g3+ymH&bvw|iVe~6<F5)Kz%F@smPP-4F`l0^9A
zcxB)T+i*)z&`xbm^L3>6Kac5;8L{B~g{)Lza*E+Y^HwMSTj$PS#(A#SIV_gd_A^aN
zi)PXTLk)NX(1v-DNWe>?I$8gf%{@5|q}rCRDOjw}VPYgUi>3ea(rb5K1TzsBHyREv
zk0&%0@Az{A9K%c%lw$66r+(V25nXY7Pevk>r$1@Lf)I!Yc@anzTIeVp<lhW1(h=+p
zJ9OTO&<=oqv(6uLb(U1k6ew;cJpd+7=M-}lUBW-6@LQP-gD9gghvjeY7X1eTz)(ZX
zEW2ApO_{mAW{DLzh8qKfBz@De;NRYu9}ergcNQS(x%#A9bA@y%()|WvG~IE+kS84J
z#{@;t{520-u>4u)<c|c#qqniUt`(AYnfHpDlb;L&!Lx3()Uq_m=+bZOq$h@#-u?EV
zkBHfV56Qpn+Arx1ARM5wJ6daBs@i6tQBC|imC1jI+xzLAefK;&;W{}Kwr-@OUW#__
zmU@eUC=0HaS}2<~Uq+fliT_(oQLYx29>XFh`)@bwTU49#Q!d528}Y{L3E)dq2Mx4C
z%1Ny$f9F(cy6bC&8P1?aEUFP*m>=f;x#2CwQ4zp3(@Lhzq)2%9UvjG|se|_YMSZp`
zcjh$Y877l<U8$r}G9MQ=V@5EjRkga&P)mV?zg9_zm)_|$wz_#0!9VfC?73C}((y4N
zfUyP*JlYJz{#xT0Zf|%TEO8FsdS*<oAqD;!E>*ARFU}u%Rzfmp@U+c9>A!A3%mm^y
zN-G;wppU<Xxqod)!j!1R9~%xTo9Yso>i%VaCYbSdYI$qWL_Yue!ve2@0Lk~6rv23;
z7J`wBEdbbz<U<3<^IroQ`HDKvwEJu`oNSYYP72W#PpFoX-Fq97F{O&MCx*Jprh|m0
zgMU9@Vmt&=Dek}79EUFaNwSpcE{M<%g&$RNM<Tf|BmMre5l1VqKp-CVuWvX1LfwC7
zDcxNNfiL`gT51n3=L;#f`0Ff~pvNObZBvJ}<sU#F{|a;OlS;t^JoY%#=N&X4Y%|dQ
zYbO!dH6V9z90LczcG|vDcY0k9k{r);?{6O2Nc7tRE7Lk5n4o(%{;c%LB&rKr9d`JM
z83h>O>1?L^Mgb^iB<k!|h<#~+WvS|4`cB8~0ZEU9sS_gU@L=<dW|A?;wO?xA>{9U*
z^WM)EM?s9^b@ZzKpMz7J3>&14z9KagfRDJ@>D4Kp>0hBJes95V2MnN21PyfWFP#Gb
zQ>bBA&;6SsNraR7-Yl!DU_OXzB=4*w$-We6S?c}QOjvPyrQox!-L>(7roA0-IQBo|
zKitsspB^baD|uyKT4)I&{I5Gaf5xk(yAQ}}Lm!t=1DnN-SkRQkS2{)^9WMj!<j)xn
zm>j3h_Me7UwGl5;2OTE@o6l82H2xC8gIH2)2JVPNI+7w){_;Kn80M_@*n1oHOf!eq
ztIdl!%;+m9#~(TlM5%0nN9mkUM5cKWf6kB`cCBA{>X{eXiewY6@P!Hx-7cPh*@5dG
z8;UDsXt#}p|Jl+AVf+!0+~0lhnxa@8qsdPVNDhZwM=UAw9173C`%)|+PyVd4BL${k
zTV>AAHA8=&wJ+$9!AXwLl;<zMAh*7@)Z#O_i=C`@)KKMHXcwO|aqG*+sXHj0Eojs&
z2%=fd|7_s46?oJlQ@@xZCyG^~<m}fq%r6FLmg1iQUnWu6$Ni6O(OW+#r_{t)%>*h~
z#<d(PU#j51zEQ{~LQ}iHH2bTr(nQwMte724z3Ad`mh|4f=RuX{jX%@p)y5yqF}vJt
zjAOfr<nIiFbc?^md@T??sBBU)EeI*E{Y#6!6XRd0gY5n>b)bC8IGHrp4j|s=f(5we
z%ePMjs0R%qEZ3s{Xj07r=XtbZm%4H@8!+{~OXF`K{D`CncE7+$k;vynrgn^fY~WW_
zWzO;L`@*2JQ`=6L8Rp5PqhQEq#Lskftnmu;w3KepAl7oN_b=fMxV^!gGwbPq9XqS{
zb^Fu!RHAVUfsOYgTxJ>-{PZ**y38OBEn*%jXZjWPJ-5)(jemS(!YC@arT)`;?Z7qL
za)e(-wroakI-~+}gfnk}Ynk&i1syF9w25;tSeRDa8TXrkQH!6~;ywh?WsL?U@nC95
zL5z^Te}nC9obt`;K2T$Z_?YCl$I7qD7p?qW#w;Go^u1-qOHrG;t8t<Onatsc=61+r
zK+GAV*^Ya!*Cdt}NDHzhrEAZ=-ogx#0V19L64_~DF6*xaUFk4Cb@tl@iafT5{&V=Z
zM?c0}j5*DEDBB^d&Hb|e8+IE(vjp2PHxI9~B;~pSTIj7Y%)h0{8uU2-``)|ejTYnY
zc)~^vxm>zpIoRap9u|YWrhS)YO|)n#-Nq}(AF#cF@=)!8SCmcp+X}7Ad42~%pl4|A
zv0Y$`7?r!}wK&zd39V@Mdgr@4=)&(wK~MG3n@hlAe$-HY#_ajZGq_X7yVKG%F}R>w
zRk6{Z!mjZ;WI{X8dd$3@f4NCM(qr+w{0P)me~q_9Y(8Ek>1aJ#bB_VXn9aw9Kw8x7
zFAd@6bk0Uq_y?J8qMwxgDJ9on#yeBYoW_jpl)eSe&~C}re!Z2FwlPJL`ab)#;zNsG
z&=$>s=NJo(*<?>Uq!+Q6?v5=uNd#nN^A7$`E&JA3!oL;g1@u3)DE8-{#cHw0oV!2?
zwXbQrD`PHD{p1%A8>uS=$5SDF;wd<RS4WkxAK;X<UVGx?eN{h9!+MG?1c3omEMi@5
zEy(g4WBd>EfcVa98a6rWI-n2m07x2pE<%qE#fO{mPerYNj{T4`aiod+4R-!RO<)|D
z_UE!5Guwfx79s4)5sRs?ec4C<-Cj87HC*5M6=!ir)=#8iUY_Hn3ELU-dTq7u!YEf%
z&bxEtdHqkR`ycwk-P>_=^-8fltEB^?Dg7;G|JB(zr$i)3oxU8tsOV)%5<C`f+IM+a
zoBsdoxd?sOqN~L?GLv5{mn|*#(*0nG*!@4-DWOZ*60-iWR7w*;_rmKyr)=xBuY6Bm
zF?as&4Vp@y^L7!Cklxl}XLRf<tL_S@#eayH^G~w<jgJ09@2K=KIL5(59Y;Wgru8hr
z3jPDMke*Pd;dbPCo_1hS@V}{CAWi;sg~xr&cqu~;1`?ZUbg}o_z2Z1#>QW@Nlu7>S
zpB}|;tNssR#LzFiu=%V_y;?K=UyJy+DfMsB_kXA$|2@0^E94V3;}FgBt{{VQv5?!D
zHPK{P%h|*K&-;P^wdMbuy0Y(@km$1xQc*aEXz$5@a=o^A)C}EpPDIjTUrLTx6m#F5
z?&z86M-4mJSp2`N7SwN6%e{?Zg(8@7$un?$+Q}r$MPZnI2Oar2kTL|5a-ACQHz|X(
zEMXaD2(TE&*`7`8r;T&tZwIaE@1V0J(v7>dL~IJbKbwxOlVEiOCp)6Y)pv%n+<Skz
z<z=+E?{Y4t5^00#Vu~#jefAYaYdg0JZ~F~44cDAZDEp@lCN8CQYPVP=H{w&iJSNA9
z%!kG@++hkdtb0HiYvX5GW+NPMBj5y+_4radrSr>Y{c@B*cZ?A$FgrnHlaZCx4saH{
z(3QG6_^&-Epoy#vUA4S-&ZszoamoFMAR{hqhS9GQCnl<{OmWU;aAx!0NbH*oX}bw`
zTW=WnIVT?#P6mn?(^V3A?vvu=l2NdvCF;&a$IZaFgEW>Y=se)zD8W8jlAEe=nlQJ0
zzV*=omq_XKW(fBheigU(<DAssM&4Jufh1PN3e>nX$kc!9ZE7f}wb*vFv_ZNfW+u8a
zD8w8dw~bw4_=Lh7V-7OFPQE!)paGeMK!ZyFy&*R%^m6n<A<(<dxLd$I1KMZTNcF0d
zvQ^{fiCxnP-Q<{QYSyosU@KUedkuQ%;LoE=fj|1(RQuFq$4$)1rAs^8O6bE53s#5i
zMG1O5G6_o(qdfpQ?a!?Z=$+!s4JohYdkv!vp7&H3=5_BI?#Up@RdVkJZCrgqWJ;1u
z(lr7lowpShtL-8R8VD4#Q4)xw+G7!w5YthtTF~Rtc{G9=dJ_g~wsKBKH#Kw2Yl-ed
zzBb36Q)a#my;C0yVuZ#h1+PPlzj0wa{I`tp+TVE&g5q*ppx>ok7@@jFY~+-dA!+^_
z(M&eWR~852QmGL7nZ=e6zDR2@m6g>&-}|)fu|v(K@(0^KdAYcSFkqi7Rm8r`<b{+6
z^QFU@8QBM}#J)>A5gt(^!JdkLR1^xb*V^g4S@v;GSG2T~24EdEouJBJPkuU>?A37g
zcbt<Rxhei>>_tAZ^3B5H#C7J375X$u9cObDhqke=jlq3)tPSubrb83sqKhG8pI<&p
zXej>v`PR%Z1J{I3+n8kj9=K94%?8J%X`09)>ZbKf7cYr7!R4k&$<f5dwOD00`=+x@
zJ-;0zHye!~zf3f#UhCntqwJE+KaTV3sNP5PYXy6AR^^d<2kqAm(T%@zq%@t6pGNW3
zHAmc^93;traQcll8RPD<u@IUn)IP{|+yK}LVfl5H$zL>ncxg;zeP;`XX^Pa+k@W+%
zOU92~Yj|Y~t#bOB<w97Q0+1z$u^a0@9(7ENmoEAdrzCDVB24c=>IO+N-Oi+QENhjc
z+KO5u^B3X<9E_RTID&~)K3o-EqMv!^^_y;!m4HFHz6>^EyC)pB_pRnKk{XQai;j!z
zCccjqK;w2{i>J?=2FY&Z<k!Sdm5FI*`Mcu#cc*UlX=<#OH3QyUoRlBiLoC{bAiVR*
z!`3)Nunh%$-Ge9|3w~%rEjfScdAALDSEP)o`>u6J!|6gYR%^9(r30TH)BZX{b(7Y^
zx$GXjr<ZZ$EwI>lz0y#ZaHP`8!Nx4{peLF&m;(2Vxt2rXTA2WZl`D%BFh3%mk!{UF
zhu$Zm-d~~wX`2=ntBKrufyTS%Lt?xmiSp0Ea3yK{C$0(cQ?o_6>A)5{W+D!d)4U55
zJ>k!UsI;7nv@=l`NbfyJ+v)B7vT#`W6koGer|FdS&T*e>YL+18Lr|7#RkJ4;5**~;
zZePj+xvsNYIUy4$pegtRj8Ai*s1f`%m>s@j=9QDiI3}`^NVS?AjobU4VB0N8-rezt
z4_VLlZV^aE`)VZyMDTGA#xvpip9Y!-WY!vdK8zK?l;LKZ({`~dG>5mGz&$zHc3sNy
z_wvg{n$EdBP4*VBT->O&#-^O3?ZlT09Jbj@HsH50-`zLFM1{(d$O-3D)ohQrvpl|S
zfFH0vZ74gZ!ySP{-2=X@h`X8_i|gVY*Ot?Vsm>`;ZVzJ`uRJ}!q<y%eqYP}RWhY?j
z+fPH1!*m-&)W=$%F^wt_nR33l=Q5kdKdOW+Eho;G+MS`}hs<_fEd_!!`N}tk#cblZ
z5(A*Oo=ji|8NrPve{q+d?FA|>&$jw6y-&MgyD2+;+*{|(X<}dDFxz=l5`i&Pkc@%7
zbRc8Sr%)-inI!W``l^F}ewCvnKF&-RUvsHH*#weIMs{r2XxDq|8F0eclV$B~HF-&9
zq*lM4-!Iw&MwLp90_bWY2p7uQ+;<sK33?dDS*v7nrp?cc5cWBbM~*!Y&Qvqpx~J{4
z%YBfI7ymo-|KFnI{QpMuF%UUDcpujr2lnqfp2K=~;8axqFkH@QuVkOJ!9i0%lT)!!
zHSQWv^{y^7tQ6F23FPTL?x_F;opsB1?u&7sgBjKd)byEUfz;l}PM-=J*7h~kB>&b;
z)znRS{3SPVC-rK}Q0;Luqm|_x*V%Cjy21n4bON940xsf<PN@JbL~FR9C+K2GBG)en
zfM46*y;-}hQVvj?@^!ImSnoiG*Hlx3K#lbzc;=vm>1_L>u}d?omr~;UHI(WXu03!8
zfKA7H4NS3i#FWxqK(Cb_NL`zIQ#OuDK2#2IGD5}AyO>kZ4?hfo-C@O@UsMirRtD!K
zYiBWRo%`p20a(H)%Io`fZmVGv<2J`uPVVE(r_)D=!X5kkhwxK@G%yM;P}@d~C-FKo
z?N&K^=C?058gRH#3FKV>*~!4hxn;gz;Z4@+mC6p)LYJBNEN=0V^99gi|H(>-DN=77
zQ1*x}zFe`3+SLE0d}4Ulf5Q`M8P~&uyL=|jqoQ>;=wz}R$h`okHkukiA4e6vmq9y^
zeV8NHn8wxotK@(s=H+=gTBWWM&*jrv){Gm4&)6^SFVg3v@}UgBFz<oCP;e=FXx8Z3
z9N}bV&U<3N#c2mV*Rw2}!CE%y9_>oC%IxiWcMjO|p7W0Ye!~xD;G*d+TVIov7ngS`
z%RR5MG^w2@?gI(e7lFp1Kwz>S%W6vk&;Wp2=Jf5ZNz^|kynaiO$TTMH?t;%SV$BIo
z&{)>!oOpQN#}8EDIj2hlD~}Qz+M@L*3Y^>II993MlpL4&_1R`oIOrtfyvHeOXy#jx
zCbn{+_5?SX(N1-*H{f(<JR>FWpyr=Clss51?npQ-=VV>by^F;zy|7R=<a9%`P3!C#
zF*PSMW<GJoSN%NQFx9loLXZnJeS<ddjoB_&SpjCMO-zh$2`6Zj;gTHRp0gAGy~X`!
z-;YJeW(s<8*ju*GTUrts*B&52c3GZCqf7XsHJ4OVm&Pe}@8fOTZVqoT{|!V+n#5+n
z&T`(&-4^@_m-KoiZ{BGsr2)tXC_MEkl5jkYl|HiH0a6p{luC?cs5TKC>)EYUJSZo2
zMKa({^l^Wdf5mwd2P84&QG0maER}Dpg@=@@zscyP$w&u@n>sf<o3)f5XFZku`DR|T
z-uXSrYdp9+Tu@%3+1~}@aKG~(^Xx;VJ`3!Za_=J`3F!L1w@^vQF_UaN+uHu<?6jNc
zqH|+weI|{{7|?Q>S+ActT|95)<k%P={@CmV+G;8XY4bRA`+D85)?x>XXm1wyW6$w1
zv&%o`d$zZ!vkdm0T6%!LCbFHeu=#4z*gM8aTjr3XcDt)FgHj^%3dId7kx-INC}-e-
zsqLqi&3)e4e$wCf2Z9bu8*H|--nnLoLy*2}<vI2E*nWTrr!DgFU|MuW!Y*5?=e$-J
z%_4fkBR$$a4@kPbx$0Y1CfvIfa;G@9Oj<>4`@{#xk(3~C^I0&}ro)qe%rXH+8-s>@
zmvkQ;H4HmlG@bG5%@l$884ar_#q2J<3**FmJbOSb3zgfB(0)l9xbL2zKqI$*Ouo1Z
zskqE@Jzw`QXB_(8+z)J5Ac_};+l@TjXx^7w-0)AX%h-14#FWaPR{QQ7wonn%g}tPe
z7}^FOd}8hcm@!m2zvd%B<R{A`#gF@beo5@JiIbhKD6a+hqpA<3-UP`X{v06A3#054
zYFww2M(XiA9xDVI;r5b|S<-tQ3i*_xz>a1#EYJ1y0#y;QZg&5&1C&a>#d?&^Nx@E~
z_@r}V9E-SPnbcM*eLK;c0vH?7Iv#g#gKGWgs_65d=vD$QTNG~3B^=L`q_Pb@XM6Ew
zFYm4+W##sYMd-HAQ*)q%aOXR66NN|6+f!WeMq7o=eoSgvWIUw+EB}lkW#_m<E5+@u
zw4A^1!6a>Z7OrjdaW}*2r8mZ{J3{UoBcw@gv&naHawd?Q!8XOr&w+15P8QYh^QKlE
zf%x~Qg>s0c?8Vn?&SF6a^EuM%!vHJ1ltv|W#9yR_<L=Xk%VNxeAFu5Z5|F`Yy`t}~
zgH-s&EF)0Zrk(!9C~#9TqB*{p1pkR3=1N1Jv3s9#p+?FK(eb@mqS|Rcf8vYHSZNR)
zT1qtCAD2mgJO(D#aX&0vnZo-3MdI%9zC_8@sh|da9f{^I3fs#zv6s)hh}hx^;cq54
zNLtR?x!U)K4;S2~9tQuow3HPTXmm+ckX_tuq>I)wetNjbBl(opN=qc<)cN#TN*ZGe
z4E$aA*4XlZPv49|)%L!_`z%fs=;1xJI)eDdMpS_J2kK4#`51}L6IOus_e>!$YmoQV
z+qKFT6pMUdNW((1sZ9(hqIN*;edq&EA~rOsnVpZ2r7U_}sF5A|(JLwY(Kgp`{U-0u
z*V4ZJ;&{pV=qTU>N4=VXUZnMQ6PYu=jjFUCsXdD<RG0#BGa%=m69H9=0n{SmS&s?$
z(xz(FweK{p&ZY$-q|e6qSA+uC;uUNQ?%qoh6FhzF2j)+8klZ}(>5EqdlB=wi+&3DD
zVt!{2>zF3FpW~D88ycC$?2M4Dg~rXxmspRnWFYPquU3KrrHq9`W(x|u)q3jNpk@m8
z9>7Vei{$#Z{7U*pKzT)^^--oI|Lw@z`W+b^5-o>1l9#WoisXe*`p5=!gm2w&t(Ra`
zt{0X@ot`8bB%J|_Y|iVbdJ8M1dp;tfX$3RnXLheCXOsh628#^Oo;}VTIGt^va+*1r
z7PX3rgK;oI_1^aVY8^*06+SG$U}O}{s!!R(waQ59CpQ{1CTf>n&g68xH+TIZ-UjPl
z|3q@pAO8;EJ@|)=+w{dqFhO#^g`e;^Xjjm~IMedQ=x2GWiFcilE$&2YS663Dw<pKO
z9hBL&V9e$LZj!lHpC07(yBf45w}!RgVMe9fX#?z6BI`n_<s?0%eSh}5S&lTEY%Oqk
z+9wjS-8S>dn(!PqXY*^$2l}FO_gbC;UE!l}ONRksHYiu&OID&A{S@pzZx2ibFA)da
z@~9e;5}?~#Y+1e?@e<o|)`4!jMk~gt{R|9HT&HfbKC7h>x4L_gmaKS+<UnUNRj6qh
z{%z(OYed!efE~df+aKZ+NKm*pE5l12t^M%Y<ufNBF!RymJ9!2m3geHu_vMWogj(hz
zMX<gL`i2EmS{)JM2I{$}NmMKusKy}0JkWOy@|hd+A|d$t{xqA)g~N*T`t@AkUL+Zk
z^vX(k{WOzweDEcbbe=ewyX@uxyGt$M_>@a3oqVDdLV{XV1l{hd#qT4WViX3kYo+@$
zE<ok#*Te1~zgz%MBV;A3lJ`Q8irpu7%XIRJq-)BziUM~7-qkNby4JzpJ_00L>&LQ6
z-{FIzYE{>jCtmcYYQ?6lBOVG|dld2gyU#rHeXg(<wcVmn$w6~p-B}@X17duj%5ksB
zH_{UxMM7dn3OX8HGd2hL2+3;VfWmuKK#2ntwy)A6Txws7o>BZMw6@82>lXT=(M@7i
z#kTd?w$S#dM`WMIL+zIUp3bARyUNf|-nQ4<HM2s>DSH0Rk3H~y@-(|Z91`?k?*jng
z1KY*kozDF+Q8uSK)h+xf=Epv$_B#;eg6Hs&D$NJNwn5GS|7vl&yG0ObYGf*ikx~z#
z+!AmOV|hgtv!Gx^mQOmUcf+|_Mrm`pkrXuEHQUyy);$h6dpPzwfs2?_R7;v4#G0X$
zU}U49pguaq#0XVTin|ceNcok~X=`#{p7f@QE7X|#m2&<t8Hmh?fBJgohc;rimp_JG
z<hKF$J1>zeB%a~?(ZSk8T&;@W@xGKTf0oH6(D=`C?_Vgi8`uf$=RWd?l<50r)%fyL
zPFjF3v-|i(#|*w@*^L91LT#>Wf3_Hp<F?(keSUyYE-k&ntw)XPoOK(Y@)9t9q8%B5
zD27kY6HbcG1cB~+wN>PeCn#F5F7TTBBu>thYSHW!i8k|pO#x)~LL<@F(jWR;_+vj+
zQ9Q&{GN@To(pc}ye+!h=RjayI0sqvm7h!)18cA_Els!2=;@%kda4{j9sJhSk*X6;y
zzAiQwI;Lj5D${bE72)Z?n`Nc~`gO+YRRpvOfe&Qkn0NsS<U<w?zvaupIxG;N%xnT+
zLu<h4-y|9LYTUwDYv(uCPuQ~U6f~mDQbbc|GDa$3zY=pOm@A%60-^D8;10+(%TtI_
z`fI1{l;={~>4|E_<mQLcc6^k2JT{6d!eiD>-e0|BSq_01$_4COv2I$BN-jl>s(?Gd
zx6S2!VNB9?ny0q<ThC@A)~Cao`xNWgJTLC@81GB(MDzj6)Db@6fn8hkw|<b+Uvc~T
z*jAWc$zwWbt7n`kRiWOX?GGbbr+gT#+nyft^tX_hbXImMsyB9CdPY&fZ|OJ68Dd}U
zjToIiaP(H=LBE&@**4H{H3&T?h%jhZ_~5g^ua#>iU0NRSIi_aqHjoS1Pf_220o|#W
z@G$#uJPY84;ahCg5#=VT{b%S+;7tI@N5%k;URToIGL6_V2A=cy3BY;q(8&ZR%f7+n
zUHw|r7XgW)hdSgz6(~7+UcHHUgAoY!Uq)31BKA(N4!2z(n0#(W_K99Qu>Sf-{6M%7
zzLrJA^8PIX6G6vk;F0ucpZRjDawfP6rn>_C`f4GSblhH#dW87ZQYH1fA7!<weh3(g
z$bFnjNU^`vn9UxweWUzl82IO9Zi(-%8Xp#~u6<F%9~S-~WMfX;vx=3R3wIs5Fu-l2
z|8(qbBCe82hC0S30RzuxgnGcE-xx`mI995zDpqmab23@Ja+uLS==CD}IrpW(Crq`|
zt|Gl5^VPSVP2&|j63Fsm`P2Ka4=G{`V10o&m;C^SEN%=lPr!N?j9Zc4r;a+SZX}r(
zebwnLWw^|2W7W3gwKv4JYj<d7Ij=qsi?d-NW;)jNYnAYXl7=K|%MAm^<1zr}#{rr?
z0l@7y1Z=1Njz)^yEmWle&<&?yZ^Q<s0%>Dy`%Vpj`qSP$*e6x$Of1JAA4&AIf4;u7
z4e-Gh3`*iv$_TDA>e5oV0e&Jytvl_#{Ut&cxoD5ur_mlwP7m%uYx_`uaSBrnY)h{N
zOhfAiEfzadO27#tG$EAjZ7d=~c88RiEIbj<ceZ_Z_DHxW*2adUn*cWdK}-NB?@Dp=
zG~Y+dyer>h_yE#8r|-`}Bdh+8{s1a;`tZWOT*aPFl9#N{t!&88Oj3*6dE?|K8Hmm8
zwzuMlqaH8*{_NxAw&#k*(z?<QS6%Z#eAm%5i(y=WAJ9xBJuI{Y0E;ds0xP8SJp8Bb
ztR};oFZ7RU?%zfy5Ujv-`3&1C!snh>9Ap9jwU6Crmi9-YRrpu6&C=_m$jZy~u@dHu
zs3AJHx&SEX`_`K(?W>)!u<gtV84b~)NW^l;2O+hP7(Kd+f_fR(T?TEfTopC%1#&iW
zi7w90X8!VE-e*4>jAl(bgAW%JnYO*2&z<MV)FL?DzL`n0I2ScSdvRcD{y?1tsln7n
z8~~cpT}><4Q-~*Gql(k|R9*?hhFl@;jbaA$O)rh5$6t2XcV2uZUbCd#zS=8{z4R#8
zQsJjFfB|lQz+sG*OnHH-t%r}&8GlI%Cb4xopU6Y7fRJt0rlyeY&<y(O$%jUcYXg=)
zzq=2R-*e;CR1Q;pZ@Mu|O5=Sy#F1;eyS>dgF1Do{s~fu4kvQW3)21ydgl&KCN5Phq
zU%p(xTE0QVr9|*Jbh3Vk%ZyCmMQ<hM^A~`a!Ej2W002f0_hBG2yqRDd1LFk;ci=UP
z@XbBtDN>M%QSV%Sztv8k1a%ILLgedsvuvI)8LRw=9que%M0dP?Rhv`&46R)F#JC>s
zDExDXU^){a;Cq*^Jmj6Yt`!%5v;Rg^>Hzs_m=WjeZW1>aHT(stoX_fFx9Aw5ZY#5J
zT|)f*UV15R4nD-qwq>rfs|e#EQV<-fra|7RXnuo(-#9u^ORiz>$7Z-V&|EjB9BY*M
zRg{O(Y-&Xo$s?(g>yfIvg(D3`pwT*E$FYwdzEz=bs-VaT)M9QZ)mKLD#uNNLH3^$u
zBRQ@A*Mxk~tOK(twE7jAcz3%uIT>Siz{yGbWl0U2B;qjS*47Zv(B9y*=+bsU&NTm{
zUK2U!Ww&!9m*q$AO_wo~Qg7F4RU(O_vL^#mPhnzA)z4O?i?xz9ELW>HyNiByve;#^
zNOKhnMm#5E+cFP$KVl%vpkFme8{Z%Oy7hV+F~j4Xx6#WuJ_huhJ>SKOu4u5z!lNH{
zuiri~q<iJBO~<6$M*L_izF#@M2m!m}pSou5tSQJZ6q!>1(qL^PreIPpCydj%852>Z
zI!nt^*iQm}M7VxJZ4kLB5LO>}8uhKh=VwmiR7^!M#2?M}j11>lUR(&%bssICWG1>s
z5JXK(lykXK+d_oqdix7>3MP+6cCq1|9@}m6qrR$B6UR%$H6)xqZ_rGy@Dag;IsAx}
z9^WN%+mBu_BgvL64fn0K?L?k{S^cVN3Wb5}3?g}`L+csh_%n4ME5z&yX?;A$OXmJy
zowKH$0czFU+re$MJw`1B2_fSq3Lm2$T1s`BS6I$BUa5XaPCZ7h3>d)0+gerEP(^rk
z_mhZ@csc7J>8?6)Ib(&)YeO^;2Od9Cag{W^48Uz_%`(0+nPA?7<!^l;UQlV{zjUqU
z8lnMk)Ov4Y#H7aKE)yJF<2F?aNxQ?&=zYKfFDXz0vp)PXzr&*QVKSm3{xxDy+|?;=
zM*U?cszMyF6?i<GlZV2$_ot-=E>3ycIRbsBcrH4Xj_^Xmk5J`gu_7YfoF{N4F2j=O
z5Dca+G(%=MFW;>J+&}_CzhFf!r)S7aEku`S%WU011LTrnz=wKM1fs#x$2!ZHz00D|
z1Ef`cj9m~d#~lBq`~)8catZiu!`=lRWyh~Iy+_aa4XX$=lDDSV0ou9;K<tuyuSB2E
zV3%omDbm(>U)wP;xAtj=T6%<daXzf(ewgZ$^v|TRrWn<WgP+|xCUi3`^I{YB89mMO
zOwi10!gJ$%rf)FApVq~{-p>c60N$BqOG%MW{F+})k5J>wP<rSqEfc>ua!=QA<ny&V
z4eyuh(i*<PiwC{nPR<^baHML-W|Cnk*x;(J6aENCKO-7SjAh;}c0ulzb7fdG%Q2fJ
zM*L{*X^np-Kx4XF>HrXjpLejrgEqdqcb*&PYcAd?A%xl!Zp@V@+hz92HSch-;oKqa
zL-{<tF9^bs3xZ1r+CU-kaqooPQp<Nu{-(>wKG3`yx4hF~@ZGgC_w<!1|7O=Nx;1Jw
z321X0QCBc+R|;krc(LkIbZora&~jlBgKkx|v#!Ta7d1L)I1}{R+EW^w-BUH`+=K|;
z5eC{UOKZiUq04kYVagre9zzdpHSu7G(QrZ(VsLwCo4By?+#%0{qU%a#YEHTdglrHO
z-KdCzc2IHkvukcZnH6OHPBMd!IW+P1pw)DDjS@O-S8ju=-@NY0RY(7)&uaSp_LK#W
znR`A#5u@+>Ze&DL59j)C!Yuj5t~|vMFfXO3`DVHdt-2+`;ct1yF60<}^#+I~FQ)8P
zdTJs;W5Y7|b@xx<S8L<2kHZhN?o#AuR~oxIeW+7?JUrbt>z$D1W%iLkciM**N?MsM
zs~X8&E;~c1;`z&lZ<X<_)Vn9ef`Un8AWOt@ekVzk`XC`&zX-ocJTE#o;iHhWE?nQ3
z%U!vcpojlamNcTwb7pRJ&kR^)F)O#9!|Ib#O~iWCXWF>=9~S-dGCHiEhJg2Q7zEHX
zGG+Dy6c*B)6_+m&F!|VK@v{*vToKTO{PJhHBJu#E3GPxHNmqQP_f#vgBKWC4MaeHC
zQly^akCx9`*J8Is0j?YYx5O>XZJFeF6FY9TaPJMn`YF8%qxK~ceq~2@H34+^1v{9R
z7ysbv=D7^xUxd|Mg7dDOtd+)dq3&OsKlw`uBT{FVl&j*&)`#uNu6~0b-@d}QemhV_
zHe~1B>)V=BdUU=(AnNJ8*qcBsa8a2K>hMyPzl}Kj=GfZ=jC$K=Nvur;jgh4FkjuzT
zHGF<BqD{RIss+3%p}ss8>4;_?C$NkA*3TpUi10-OnO+9@z6}3kTNk~vsH@T&2lEep
z&|~Y&jg~0fgb~kP!Cg=i#ff%S?%e(xza(|%;E<~moGlrOjvoGqV&s|i1ih)9BN5v^
zr-Kf%GXm6VVN?<JnXYzb>bk;LMu@^o*}hrsa?ANZxb|C2!W$(vU32)^G01VDfWYZV
znHyD^aBX*+-BzzTKHKq!a+O=+-nx_-jlC|4BU7@WsA0Ob*g&s-F(J(Il1d*tJ<qM;
z=5XLurUErAU`Wx&#gyuFrUkRvHF4Y59OsbNR@dkAQ}xZi?CYTW?{XH%<-N5G&42Q|
ze9H-pEEG7(T~k+h5kDR4d^54_YWz)F^pPmxJ@F{u?N^Vmlf+z`9tchrr}gw;x-y_(
z%1h~r_Tbecy<67NpYtCiVdkLJx=~!om=a~g3A47yRct>{Z_=$_9s7PG0z`gEUFlxp
z>mq((!?NwEe39XUT)^YlW_sv-v2)ch8dOQ6*{6jG+!qj%W@3c2D!nr_apWzxY!5s%
zYt%+;aicAJQ?*G#)(gTFAlh-9!<w7ZuurZ$x`dqXuE7ij+h_;l0ie&^fh+rY9sNS*
z?n?P>@e7E2zK|G(2kCQ%Vj+~vPXq*|N%Yht$?&NH+zt!*n)A6#5#doIiwP=K*EPqA
z80E0!@vn_~Y{v3mL84uLxIhzP!v^JOZ*lYFQ>bM%>cTSU5S7CU>U_@1N6&Mq!gmGY
z**31hM@W5`VuYSiuWP>r3|=u4TU~S3oAy4^X%wD*N<J7|RkH*n#;3L+1mX~{88J8|
zjRQb;CdFR)k#nONdtaqnWnVpBx$0Ufjh3D%JDPS-<>)bYe?bj#f>EB#%gc9g2uA3O
za(z6@;4~qd?hA`%B-zV}uT=Lb0pLY2JW}^qQZ%&tFgF)_33CTjYQILH67us?0rL6h
z$5jOV89M?9#uzv$vaaVb5c(WeKh-F7PrcX95h2P^eZpL{lAD+WWG}SPH(s(I9e=c=
zmap)c^dE&6^De)P;QMxsUds9lIa${Vh4J$j4`1)f@u^l3%;=Xq7$CbC6E3*Gf=;Q^
zew)9{+A+bo6pegtdw)?`{De84d$n;LVpvrjO`VtwWqV8K<jl%|=f1Bl0#xB9xh9XK
ziUrjH47?lVupt~s!WuI3{Jh59{kWu1lMeF2eqs9w4gt7<hf7kQ)j}oOhye%P2)qk)
z-YAPz$UhN!fk0bGKq_H16PBTU9b4l33@R}m&O9oy6y!^n4{t~nnfwAkE%2gp>{l5&
zHEx%-*L*k3Rf<d<h3S-OpvE`I09`0BNcAvyUS8D#&cqePrM*-EB@*3?x>~(qGGx;$
z$AX_uV5Qg(;517eHQspp%kh0ZLM<}uybUPQR^2?%tu<P`1ya!1eSsyI8qFyBd5>Uw
zT9L!hd;uKt=BfXI<WPm4sd!Q0+5HzE_LBtAFTRit+O>2=AoJ6K*8#k&a>`j!wDNDn
zPju^LC>gYMsne-lzm@NqM$pqrJaZ+>m;EMYu7h=7G-~@JrkxSm>Rn*`LPofW4Agpo
zO6sSisAa24q5poMZC;EQsdPE{1J#a1iu`>epI%CcF|GR2Abs>{gS~HLo_eb>y`0uP
zWobd+g<MN91tZByX-4pfNtK&>0@Xu&c){%uN|FR+2IXs0S7o)x>`Yf|DpLFVSzUo?
znY0mmU<OtFMk8ggQ68l6!ln=25=G{$JfYe0$W?mh!%oiuhXvk%<i!s~pz0YG$Uw8Y
z3JY?t+FohBVPN8ryre>S5!Bwn;t)h0Kd%R2zA?eMZ^YlWl<0UU42^rA_hAwq64MP*
z*G5G|B6+J9Kh2WQJcwcSVx#=PD=YDYyVSD15I(?seXn0375gapmVPA%qm!F8<>zBv
zo@M7KGSG$<+;;<%<HHPPiYEB+47`ITGr-1{%t~HS6KNzRlC}Sq`^`SJVq^XX(m??x
zI%lbL8(|ni-9{WOx3V=~=*ApuLXIq>zB9m$ei43q#h_S_DpM4vMUpC(Zu7rLy2`Mq
zx~?th&<rij(B0iQ#0*jc(lLP29U=lsNOw0V49X}VEr@_aHxeQbh!P?VN{8sTeZOD)
z;eu=LbIx9C-L=(OC0ppNltfr1AFRx2=E9QEYxaq1GTr9)fCHhM#i#Al9c3RwP8V_C
z{6{a?GZ&;7`h&mG-h`!GQJKUktSL48Vag_<_MN(JpP1##qiuN&ub=Y2YkhM79WNus
zhV5YM4(x;KKGnKCQh-UU(-d4>Tlk3~#*WK>I&>iaSP^qjjV8$3&zSq%R6$V=W^!#@
zdIKx5$ubyqd5+n_^P7EHt%z9sCXHY9bPjbsJ=gj!Es$}VUtf`W2U%iCu&}#43<jnz
z<t5!7_iw!BoA-n5rw&lvkKGRN5%KX7lZ*Yr;`<)K<nwZ~h=^{~-)}u~86J6guI3O;
zVW@V|v{D34BWQIM+8KlB;~5<<zgw@laH(*mAgGh>&L(tPa3X0jTexzJX;lC!-R019
zaz9L~2XGnzdd$D+&!uvK>|EP_K%`p&zS*4@4&8_-Y9;0OXWrlS9*Il*iX2qT%+q@M
zb8*ArfmOljJu~sn!Pw?98|Ca+Bq7Ub%rlYP>jZ+1AMD5xC4gDmxy?)=T-+->jYm#5
zRTPOkZg|LI<NjuC&uD<g&cY?5V^Jib<!fhHJk}4ZN&#U;$~-hyzEyvwuJ_mLznbV&
zGZDcl5ESwBblf_6`msCOSYu91Y*~u|ZoSP1JcUYcH&fgkX7AMF@uyzCkx92f;BV+r
zkjyN!7u2X80|?`o0pX-YwACL_YAqN}DyNhT``q*iRX;^CnA*Au4n=6|hJ6C9X^=uf
zbWM78fTLOI4@ryDs0pi^?BFTo1U^0j9yKTzvrK~3J_iwD+Q;XvlgVQM@$-GFPN0X<
zyaRdmBKA_cf+|St4y#_VcKye#%)H4<HGN)Cxxz;fj<lsNKZFQ4uuHKw=eb|)UOsiN
z^voDIRXC$7(f3vZdyzD8qeU;8LY>-xJK7)yHuvS&tvy53IZ~54?y6)tvAJB|8aHsA
zAinkxdKIqA8}Y^<CaWRaW8ubC{#;`Yp-m7U*2n{F1=;BC$M=0y!S_>yn4-v;Bskwo
zz*!Zr(|@eEA~en3WSJoTZNRf^%5r(@%Af$678<X0oyZZW*M}M{n8eiH{u-tcez}Uq
z3=PTeOIj$Xd)QSh>YJpYh54fk<=uC8>U#}I`u-bb>$8^rOGfP35vvt5Y!YNP%S+SZ
z^1!+MBvA}rC{$==gGJvKI4q`4Lxs_y?x?`a^t|(Csi$r!Il^$izc$O}xO=7Jl(Ow!
zd3$zUgI4+w;vUIMBKNFyl%8X%sKw2|#{6fgZP-<jS#3lX{*<a9FBt%DzUgv!mcy4R
zm7uTJU%aoB^_(h_OQ3D%B$49`?2q7YSx1Z$u~4JF6Xq&cq{Ptv{m0>m!4T)P1WvWh
z$hojy$&0es?>6;#TQ_xn`Aa|c&=v(<_46;xuK}$kTR5Bg-z9Ge>uMr1hx^GMGLt<p
z_O?h!>=l^q@6|%bie*k^!wDxf(@0z7h{+YiN=lP`hp#RM;ROjYQ2#72yMRb5LbIY{
zjGbHIFd7vm_*1=x@Z($GZ<Hrm{+O0d;WzJuN}BmvE_&Yng<T?^2)6W-$jNsy(tGA1
z;VkJ+IeblAMOH(_nyC>5?-s>K<cLY&I*K~PwOIzsZxrWh9&!0ZYl4K4s?0S)7TMr?
zr<hh1!Kdgo{Cc!&kd98-@{PjrSQ>V~<vtYo@^Q+267f?bL`n4<QQRkLFO3L_N^xwh
z`K(s`U8&>-TYzDSr;VxMo@EUps_<?O-d=(x!DFb2-CcWiz^v+jaIY}77(8*$K0U(4
zze*5U{l9;7<ELa{m{74SG8H=lb?%lbUMgH%HxF}BPT{ho89c@T>o<O<j|+RxODI*{
z&@+tuj)K~^KMVa}xhH%Es3-*-F@zpo#4{(do*+laKwy9!kIlkERqg5*-aG{bXmLWx
zr&aaxmKb?KqsV<a{yC94x8&k#Jb-1Md%L49bB!T5VkT?L$vk7<Y(@yYrwu+Lkauqy
zpG@`-;K&lM2t2uB-TS0^2dt{|(3JxD-zG&ku$>HMHuY0ncN+~Y+oQ*mUhXe-SiFz3
zF!vL}+!fFk2tgEqr+?3CL(F%2{&Vx%gw*Sr!xVuzOZ|ZPpZ<<7b(_l)lzT$DuTQDS
z3JT4PeVEhIm@>_<ABJw%<3*cUH{P((wtNxsF^uNR%>*H@tVim9ek2l2cjbw+4*E+e
zWG;j+&ph$BBT>-v&fuA^7I!9jrO@CqEVPgicR!Sz<#XYs?Ne?%0zv`&L`}~;@|#sg
zgrS&LfvUbDqA48>KKEZgd2l*?Yz|^A>)MWlEYBJfAK$%#-c`}MIry~cUW5JN0bTz>
zhA?G8HN2f@*$SJ7CU9Bso0xgQ9d7E4i>s_|wikGP?l+T-;jOympoVjo`t=l#v=Rv#
zY4>jmLjRh_7hYe1*>zgyc3cWs-+wou4X5f&`|qU3DddB~`<TwCwuJKfIIWRMrJJ&b
zC6mef;+oA|V)WUuAbg|;TE02a3a93rrPc~1m#S#-4F7{c+<Z!~*_LC$u-H47xM}XG
z@qOk=W<ceetOZWP`u}9y7IuxTrv?=7FXsnGMoIJ@(1>tCqQwRHjc;RVCu?&TMBMv}
zV)2+fp#vw~h-iwi6O)B?lmL!+3?By(!md<<SLZvEgr)LY&Qzdmeoyra0HQoDT)I0h
zV8r@~&MQ=9M=BXznMb%06%&v=6Uy{J7B<QNkF^Q>ZWxrHPXx!B)&IKM(s&h7{<7r_
zhsv3j-`{i)HsnA9X{mBK`Q$4bak=Z2R{Fr&GQ&ZLekAnXsq#k7RpGZJ37rf?I?Zi5
zuC&8y!Eyb1ymYn3CQ(F$6F)|)@+y~rtyrc#>MFA$;I~gRMdJcH8TH16#2=pU3-PC}
z*x5>+`7cE+BBMqasP}YFqrYEyAO-c+?EB1>igf1;tx&>~tmCWWPJfdgfqOUGT2U<b
z=siX9wLI*L63_qw(Re!kvX&l@qR^Wja8Rzb=5zZm=Ilc3r2UpZ{EXvl-VKW=_aKvF
z<&mGRI@(Cy`-CZqP7OYuRuSV*l!;Ca5}uM#@<tSJ*>ASJ;<Bg4ig+;7V^gxA`|z(|
ztVH&ig90L8FkR^qp3&bWP~;N&Y&bieR4I$4M?hIHXoAb6SOgl<M<S>`T`lrw{0hqR
zB@lje@hw8^9`AR-`603T_vFFBcf^KJM>eE+clBPeX;X7LRi3$-FLFP+(`V{2%Hr^c
zj7y{0+@lHT=Ypc~ysh%e1$Y+<a8^Rnx0PcU(g;=#Nd1K8cRh)PA;|sYe+J=PUl*;I
z=5P!124ZrW-2g%VirO6XZA)6NyoJMO8Gy1_ox1DJ!LUoFgsSjLNr1XD(VC_}eVOQG
z*%88%`T8wHEE0>ULG=ecXDBoM`d>rkDoQQji{dwc6G8j_BAy4ZpU~Ld`g&8uu4mKS
zgMgJXSW3Z?VJ_!13WV{bF-U&+pyzR>sd(CdonRE;`Hp&J_MJqJ_}d>T6Xa8a&+2v4
z*&T&)15!lX%@S3JSh`c`pqa8tlJ4aJXpIb)DEmJAjsDUEle9ep<^49l;Kx`|qwsU_
z-#SHR#lrIyPUF#bIj@`p6KTEgRcw+id5e+X*~_LDxXK`oG{ODQ8dqqF$NH`<41S75
z-5jei46#vkqEJqEu>mGaKwMh2i)<*6#nO498DG}k^`@VqWF;bodiQb)%<>x~J>z!L
zZ2Qk3@U9jpb3aQH<dnRzoy=yp>(>S8K&sSE5L7lWe@ZVJFDi9eZ!h%2W~P<{R+N;I
zAKN7g+U!u0jb=00yx%rF{L2mkPqx&ZH~6hcn|@HXj8Jey%JGF^h2QMfyo?97`X#S8
zPUoc&mQeceeCb@(y*uEUw80fbWZz=i?H>iL5*Of={toLQ;<MfRB-2UXbj*jy2K2eD
z_gg3s&5|Ed#2!wDEN1a-(^O6}rvkb{<Xroz#+X=l3L<_+UNO-nrdxyeI3Y~&ohmSa
zLF6VRTZLNgA{~iXZ2!r=dS7d+vJmIX*!1;O#}_&>(Kgva@fB>bhP<Mn!O=K>x1jw~
z4Aj8T7poP~6jNzS5;1~1F{if>NGi_>0%5sE%E;C2#X}me3|$W;(Oih!AHJ$v5hWCY
z1{2V~9g8#Qn^Y+4n^AE`Z{5v^Z^)%4)n|Vt03g+NOg*Er<kN#OX?)v>+dnQIv+QLr
z=xgq|@uo{7W2pXWCy9=8GONR(k4r3@ctj~^;)NIV>|i5aEN5bZcf7Ji%odc!WTMAo
ziMnom@Kdv#?%l?Js{UVOKtHoF4P8i@-Bg}VJ>JH=?sDbFXo|KQAm?(7wx=D41^lh1
zC?E5mguwM5!tnLuC9Y)mHkmvG=i5lyB#&Ob98U7Z<%KkK(0f&F&rV0CQhyp3I-dRz
z-LhUsnKtkI{rx9%o0d86b;0?=g5Q~WNFtU}e0*Ug#TW>S$qo0MGQvz*m=~(}sX?XG
zDCI9LZ-ytdT=nj=;P-kGYN9&K*-uj=dl|k&j7HH}wGR$v`&*K}EE&AD8eir1u<ZL@
zH?_Sx<Gm_0dzc=Lr{9^Xeh)ffHt#9D_q@o1i5uBC2CP1+U^*;oNN8Rp=YaY|TV{0z
zQcxdf)9#e&EFPRz>*oBCr$u+=ao+~gJJ<LjsABN*@Ex1<#J$kl2FyFAyq_*Gv93Ce
z3gV>-<->ZBN=YcE=ODRAFREeo=I1*eXNh3_X8|-KshZnD2Bp@_)m&p8Zn?9*wsibt
zr;f)8$iCucqjs}*?~&F*^j_5bjnvZcg~a+7h?v2|ZjpHzA*e*!jG>Fh*nLDLey~<h
zxUF-V=)#1COuFoNJ%@Z}9{pS=H$|Lp?n3r*#QA-!DAnPq17;3MBB4=ZY6HfHz$u6o
zLw$?S(!1D`_#CpksLz&l=qB`H4j<R7!ixX`WjQVWmiC`ES7!(1uY3m>Iemops4D2}
zLF&x*e-bxJ>=o0V(XlF<2>aw{#t_FQl?7#gXYv%CkHlUTd$(F?%imKx2bE5IKB>)V
z%5IK!r0F!p$*ydKt0k8p8&a?&0)DdT9enK^weA*8Lo!Mj>a$0S3Z|04KTsRODXShN
z=h)ZMg)_+-LFANKa@XFv&nozzC%^qhQrV#raogi2N+qERsl=6r*O9p}t)<@cEPt0s
zd(X-%ZnNyoC2R<^G+i^ZOaB$MRG0E;Qdln_m`XhPdL%M1vjkI~GnLbS+(DI%GKg7S
zE09F~9f=gGRsZqW&vO}x>zYc0_>Xn(HrA)-0~@UvA&YS!yg==ZSstNQYVwM*?9rc_
z!nX&9^a@-{8Z9|Yztk!SMWYES*xh@7g|6t=ORhdY8d=NK^C@NKI)6VSLbqdnse&L|
zCEWmL0{)rQ1SB(dGlQ7j3i}=E=spa(G{NVXW}2PmOf{>3gH9*qQ>M}n8H`go0&I^w
znSq?@w&dV9FK5nSjW}mV^zP%dzW-dyO*792yYr65>>1y3G^{G<Bh$|Ei%PA<LNCQp
z9+)|dT(UCmhI)>9J=e!<mr(px%{`$%F?2jwy{AwsqPH&$9{DnZ7t6e6C8oWc+S997
zt&n08^8wY;FfB5}^OHzU)%sOBA9v=D20N>27926AiRqS}cdiy`j?A++EQ*vaI5b+M
z45?SFL^y2_Dtr|1;XIMY_=vOR3Z$ovu03B1s+&@h-CfYNBj0+dvJJTfeL-^O+W&QQ
z9x}nI{QM|oKc9#g2}wdKAM=27EoJOt9HEsPR*YksOs=}^zo=Jk;gRkSsxTca2LA5+
zam8lmqZ86=DyX{B9}J#wDfOwt#IwB1CKQZ|V(KwrL!5%4=aU4teP?Vf>hRD3E(wL3
zbol!NSicD(mtdny@om6$>RVPo^JnJ2gPrKJ*JjOkVe_^PE^mJYk%tIMAn`4wZgMca
zmC^g6<l{g8t=zo1q2x}P>~JveJS5Pk5n)=l1w#*z2M2^c@iCVWAkP>K25xY98JQBd
z7HtgKfO)NJHNbx2oVlZ^-{u(!`dNzb>6bEp?lmB}pz%^5tgy%(lj+H;Mr17r@jepo
zIPuPfz=;i0(yCfjhCrjHC7W8^YGl|@TA5mRzZ9FFhx#{~L)|g-itI}AaGs{0Ud*}C
zK7y&ppt@OD9nDx@(`p3`=5gO2c35*2Z01XMS4DB?g*fppW)>rtkR%JvF3W-f2=kL8
zwGN(}Eoc<qDw)H1g;n@Q{(|>wcr!o^f$%TK6-}@YHi#KBn@z)Gi!0J;ULnga`&*Rz
z4iu(&wF5n-N;L=Wx_}P?)#4f&WkH9-D+j!X8qWjmJv;BO0AnUu_cvFih1J{jT-Fux
zA=cHVm96oHNE)LqLKaS%#0fp`s%P~D>r@zM5PRR!Xb|TZtMifwyrkJkP^jI$3c$!y
z(2u`fS`FM#hQZ=*I5kgBOHDrx1-Xw<91l$UR@^Kl1Rs8Dbj<G{kgDX-w!0%;at_m*
zs<VDv0CT7)W6o^4@toG|)3y8s(F7_Es_lu<Cz;6PjwfJo02?7N2`4mpCZiv~`4E#3
z&|Xo?iems{8>AKU&Kk}vn=`w{WvH*}B$gyR&zb7xe52^|^0R&L!K-V%G&ZzUBXxSZ
z_9~7EL8i09e50E-ezF%6icgqUPXBqd5NRj&g=bxiEHm~!kxe_axOoSALFZ)~qBL4#
zNJM>x;66qQ28+)$IKPBH&zg>-$<={p|3l5$C^H2?7t@_=n2tAp)gvu+1Vhv}Jf=U%
ztm1>iSFG#9@v|8>zuY8eqJHoYUK0A<9GeI^Pk%qRAbdYe<RXfii%7VIKn_8u{MtJY
zM)fzgTN94>2V4U4d0mxVmNUqIAxS>2y0*21s9?LAS*P;u*5}fAYGiPaJBum~7W;OE
zy7eA=x-<S$L4x*HKoIswu;qT07!eEGtx~@#=^8(4%#{bU1ql<HuaM~Bb?62<-8fl+
zz>Tt<D#PJeN)X%xO*E9C0<i|>Kks1^VFtPPRb|VgDMp|rDPvnH;ShHHqOe;x!H~lz
zJ${6+JFHgB8M3K&6i*n%SaUO?kRpj15gF+@tN>He)V>y%Z4AXY30m_9Ct313InGpL
z=G%M|)D|KTZ()aD&MdT;7rCnjd0H_2kGIqhZONGqe>z}3@;oWynR}Llb~dp?MZ!7v
zvRKbgWMY4fDLe);`DE|BnhMcmv_4=yIbtBf!Ro8|Oz-Q}gCETvWx5N?2AYFzif2KA
zf!M?9w*zai;Cy6DRU8Eedz?bmDe7b4RAx?5tSK3iM2=C1RDqq9FFJ{inR^Ar{WSpn
z9a&rcCXE=Z-pYk7z*KVj9&SwpJbebQ$fiE5o;OX<T-Tn`Xy+ZF!W=1g8C3MNnn@e`
zd=PunBfCQeFO5gp39++3q75>W?zOV%`%-Vc`kp8N4m*|&jwz#!0^OI6T%_MAJ$?>j
zOoj<L^O!^$2Ix&nR7}6Pd#4|>>p$casOkxLDxBG-5^=xk57{>%ijP2YmO;o;$n8a8
zG7y?^sXxM))d(TK0lh@o<cL&=)}r-==S0X)v!<`V5U@HQzmrxj*QAI}@6vd;0k9gL
z#8%+%kjSP=uk;~eSr&Q7YL+FP8uK4FBAC5}Cm9|1JkS}7m*Nj{{8pM>=Gc(RmqrBm
z2|f4z^%XQ;R&ys>UzXb{(;bkvL0o0}&*y<gfoupP$)fCqLrh6np<z^}*TUd~^-dr5
zb80V0h0IR23HNtOPA}10i%Zu+{q$b&wT3+vrDOp*U)tKtg-A_F>E%KJopEaX604@5
zxLI&o<$!V3iFryO5t1i3eDvF5yIZeR+jadV)N`3s&K;4q`G=vA50T!-&6%O?_mbY*
zozMo1WM9(n_<_IYD#V|A#(BIq&8ZSFYOu~v0$@3B+uM)hJ8D_Ct?Th7f9JTw@!<Nu
zIVFoqg-m>^u*Lo8sou=9@u5&pUe`*Pmbk1vmUkY*M6S-|zsdo)0<;t*ypjC?Ez`d{
z5WpB&8;2$+@hwt<C>blcgAq!oJ%0~QS;0}o$M^mE9x2EmWh&(VkEQa_^sa(Wh%y1G
z2e0dRQK=1EExi|rVA_|s-VVI~mpe@XR^yws5LulTc3nNlb2xsPmT-5wx?0f*z4gK*
zTVU{XGYDEYmJ;;Vv?C`NaL>8t;$Ba2dCC&t`QA{6f79_ukIL-e(G|4J)YxXB{)Ccw
zMduC&aS-NdmQB>xsjsisrDx`9wDXB^_TYqAg)e66vVB@vIkCcq;o`E_;<)+6TXdg0
z|G<*Gd2fA5Tul$g2u7{Hz@An1(Kxi3`0v`-s9kCK(2bTSvkiKCxf${3N;#-2bfaeX
zDgHztvu**rg^)#%4Hk#szK`NafmF4=A6~O>rKK&XWt7iYw&n^tq>=6-T)g6_A?Mcf
zXJaw>g5f3H-5o&T%TXj`PgUy1v&omnHv?7BFF_yK$4Z`kakDPCvaJj(%X(RVUG++j
zdVf{#Xp1ULe0Ey$+&_X~Rql;0hxuwZwIya^cc<BjV)35D;D-{3kK%no^8Kopw)OAz
z7L=L<s^B(R=-2QXyw?bc*NX4ho=636&r&YxJLVH_eFk?gRB5hez-_R4F3j8Z^?2>B
zgmCO5>JmZu8%72otWE1H7=+Tw^;n?tWsM&~Oqz0ADm6HpoF3V;VFAuKAfQ*28Q1Hw
zgl(oJoi-N55kK9!wTJM=lqZDia4U2T?br`UN>}E}?yvIk8FS}-R^0(*zqMaB$!9~}
zzYCeV-GHd$q7noelMg3@T@e@7BvP!?lC<r70SrrJgwXJ^_(X#YC4QlS<?RJe#yEGg
z-v+$hscMlVHnTzhc_KH8`2w=ykiQx6`@eR(_c5!b;T73s#q4AZ81gbnK(&*6Tm|!x
z;m9mEpm)tpl1-+Qy%%lI3k?g8>X3oZGOk`~Ae#N}<dRn~YXPDn%_<uqsGh=opsRfj
z5^e*zj?->e%~9Y+Q~UKO@I>}FVvF0qOj)o31qkfc)B^l5f3S1{KKtz~+4tpeSt1tW
zF;UTBjZ#wh+bh^X*eyu|W*eFra;bzqli7VA@tft*shU)S9Q36@Ds<mE7FU<QR$7)P
z#Lt!yR*>KZQQA{sANVoLmD*dgc8AJeyw{_jfu>q6btKShucFJ!92;^mFkR)i#3ft)
zUWe{QsEmc+$ce*m9hA%OQl7alyo!d3_gy-wN`Kx4s4cpuNcjC|s*W}QF|yF3`Zaja
z+h=O4O~SPY{fzV7iIi)B*?Gi?NpdcUO~mM+?q-=3Gyyr&FRm6z4`+t^C?r+@05k2J
zP6c)&2Fgy~ha@_Ci}VrwD<%4;{ZYG}XDdK!b+g!K>Cr#0n}lmwsQo;N{((L$Il|;`
zliSodZs_n{;8=re(Hr&iR7y5wcloE*ivL~`0Ds=@$7@^k`-#!dX6PJaaX6DQT&jDr
zL2CL`f$j9<^EW#C1Lv3HQb1@+SA3e*`w~qs%{z7L9fXlEr9+fN>HcHkisbI9h>H3C
zT132KHUJkjGS7q!3rV)B!9PQHL>%Hj*Uq~UHR<a@eOoBQP%U}#c`KDNU!#6=(mV0p
z#63uL0P%zM({#*;USWTt>~D#yHvNFMJwO#y&h|%0k-BfczvMzR17&rY6Ger|?U87q
z)Y*>c@#yi4i8LGLK6>xmq=7gZuM%>l4L41d7V9y-)Uu%p6d8^oHafg#P()1cd)8pC
zUjMGN>jX%=RO<`cyQBwp)Iq$cm7t4ME6>`%@efffRYeq=fZyx-yL;qveezb)v+m+z
z`O2RBylm^!j*Rr9<F=3P(AHL3(!*-9q{@Y83dkTB`QHfE^Y-lSud#ePlcJ^d`+0h)
zdUU-31YR~E#iy<7<k8uFMA0p2m+%*cBX7~pIpggOS~R*7z6tyJ#7ZuGdME=LSn1>-
zaZTInSfG-rY+Ch&$J4owC0Fm6IP|4B@riS~o%`@LbM_%W1x{h5{iAUJJ&rHN#i&E)
z|C^2Ej)~f!{PF6wFWomrUF$2Ywv@#+h;(>gANpfQ_kR!9?0U063VWu9a0^eoeiabG
zeNFhPQek3JQ87;M6`CwdR#Da>a#4KMHqDFW!TnejqtI#{v8NJVuw?d}?4+cuyLt$O
zdw7~pWF>z>w>T32GX{l(m;9C#W(oErWxmj0d*9+-66d@vq$AKd(`N-i28V7MJ1zyk
zrBs?!C*+)601g4?C~4+w?1X4)#ELO+cCv&IUa@rre!9ycue@9TWb-?ce&>fKK;TSI
zv3t*(_@pu{t_>Jo`HK@kQ69WEDOMl7+4bwRu-R)s(?wv;?Mc1?9OKDCT0G^etYJmC
zUE2W$$7iBUXeZMTo;opH+Oj0q&El(mZd*>HDw1;(gN>=2NJwDCs+NsJ3S>R~4osaa
z=Z&TW#h3Zc4u45oOWh}wgs*9D19k%#!)~3wA;E3@Z9M4!srLT$-k<#jyz_V!Z4arv
z;PHySq3}u`+bI{KVh{pONN(wx*2S3~R*{2lS}v~CkKtTsonM^+Su<faZ<ICDx+09^
zk{Mc_=y<YH@Xtrmu?eagGKYF~$sepfET=Q5Z#Vw@p%SO3UGw&<P89BXNU|&Kc+=m*
zj{+$!<N*>x!KZA(ImXV=O?7#u&ebfx)fs9EmZ+dqM^K5ji8nQP6)fKC{?d-x_CUAK
zlj8cyNMCiQgE@M9&KG*Z)gri;O$0Ls9#@NpoqMey+{mZk&&%1~B}NeVR;`MmOjc@A
z*@0K)IERrdC0a`Tm(&%dM<X4m_yn@88AA#2+Iqs~MFt$#GA)~i&JK&dPyBwicOU#z
z1u}APxkaxx5-mF3{&SG~?9mhr`azK5H!{rh@$}yJdZ=G?^%cc>8W1X_t8vw$0~_>X
z!dGd-3ig){HvN_yH$V2GK5fcjfU3((K|g8B8`5eILY|wJnLyYMAG1SPm*<0}y&J!z
zdB7vdv)R8DzVE>OZARyFT}ZL?gn}}w$yMBt=#&_Coc~V6=Hgr5wu!vieHJ7+W>0NC
zXvQBG+_Lpx>(RIS47N6xp$Wzm06+LHFgfH&Awyg;D={TVi^mnrAivXW9!(KP$-*G^
zRTOYX?AMS{<pvE={RDE3280A>n@zJQP_DlIF@5nlshqYdCnEDJhs>tsuP}5k1%KS<
zbn8zw)GIved@c9!a6>4N_-=3fZOAmRo{<<ftrm)VdSm-@Rcj48Z2D9t<<?Q*GV|Y#
zUfKA&7IscP9jc*A#%`849j4IQmO0X+VUQ4uy@}J=ZZ|4t(_ZtO6d$WLQBEl50;W_p
z^Ll?H#9kM+3w>o1zvCrXfnJ+Z5Gm$v$t9D9A*hbp_jf-C033Ig)VFzRph17~yY|=X
zp<n!Cq`R99ln*`)gp4Oj|42GWMU8YN%+_}QxwE2NbW7^{2jiboE5gDm9+|xQm{t+$
z_H`wscW6`sGayJMHV>d`BX<_~#RbwGQ<WNNP+<%YJ>r{HDZu#l_ThMO=ttWp5nC<y
zSj_7s7c&c8n6ww}rY383!sj2IvWhX;=e!I|ua|uZ=b-6&4ZIrZZ>)=`JLYn#9Sqg4
zduRlwob~H)Zh(v9UWl^$-ohzd1yQk2ypLBs=$Nr=sN#>!MfdU{lxa?QQ-$}kZ|2Uk
zyX)*TZZ{S1;6lGR#+L#UCYknPbj#djMf*wXv8>MY1F~=G0pGiD8-HH*o=HHK&@pzs
zMR0cpH^D=_#4O_euNO4|49f;FwN`7H774N;uxP5EnRkjt*X+N4(Ob-3E;>@E@oxQ3
z`ls_G>E|WF=)>cAe(5$D4B7@*YjM@#Gaok#v6|PqbB`jeQTK3j4RL$%Y$=mfVP}Ou
zPL-z>l|kOFLRA~$Cf)Ocwza<5$=5YG){xc-xvj{aOU;-ElywI;drto36-sbjH+jG2
znipVw$rF=k6c)a1x~gXN5Ck^&?u@szE+)7@6*((tPW~k=o?G+jqO)iD8WQT89oX+R
zsHVlQCxOdItSoOBUNcLJSrPYN-gNm)$)P#>r;_rtztyq#>`v!b8|JIa?Q5z+wTKCb
z+ST8Zg9`<whphLzuKP?^iUJ{zqbaSzE!m?}6Xen@E7_r$+<T@Z`-B9S3h*vvsOBRh
zUcbEtZPc<Ahod0S?)i&)UEo8tFDIbfXC!qe9TUOwE~&CzE?`*rBPbvSPPWalgjfIZ
z3i!FuvC(9u?Aw@u^mx5)+wfstHkU-HnA*@hw?TYs@G;ro{*(_z<O$T*3_jCbB&z(j
z6MgWAYq<kB`c20{z612D04vALm6iUG#V--S#0TfySEcrNU`16YKl-TI?%-w2P6QI2
z-ETcd@X##Vz$&12#LkNCTfc~A9zh0amitS$;@thElR<C;K;65`?K!L1I|^tQ7sVZk
z7hJ=q56L;wO`Q-hCFJS-IWfWQwmW1S4|<+P5a60kmz&B+yhaVx&L5)#%gOunFL(0i
zx_sK#eVSI{%B8L^OF6<1zAT45Ae~qVUTJBalHD#VSQ)xwL1>6gWCb%yg$dxclW}oZ
zwlm1Y3F4OM)b5BxIbuTy?;n?z1d$ysB#<GxQO(7ycisEoex-g1LpIC(_uUjHTkbe#
zbASTTteR1O%V^7rH#hWTa?rg)5^Zi%ukDs|4L>Z|n}HfTUDx|y8yg4$uc;_mm$E+K
z{!^-SAY$9G0{x2o1nc^UlLMpyD7Sdqpz174*Sp4jZDbp(%MW++B6{#nq1P!$Y%97q
zQU0)}UBuFZaU$51P(=R2@x!x6N1O*WpJzdTqDYeVt;?m5EacZjGkOtDXbG#Fjl3BZ
zt<l2}YUhp9IVQOZWV~2^99W)Q$X~Z7sqr@5%UKzg`VQqq@aspHQAImb_?-eVZrZ@j
z=lsPh+z&{YJo${up!Vup5&l3~wr}xUM7pHm7AoUB!lD`PPYQ$!KWSsX>o0EPLAf))
z!=bkZzV3Rm-(;|xX|}C!q5vw4&}2E{NBgu21uX{fg(ZGr|L9cDVIBrz&pQf9d2XC@
zULPT;S`-Rhu$`m#EXaDt*=p<I#HKW8#rWrHaU<{|P|Hv8v-zoI6EbQQ{3rkP=h;$7
z($*)>1go&Uld+G_Z3=;1`&mt(U0s9~M%Q$u%}OJ)&?q$d6fngVhN-{ZgtnWvPo~T?
zilnSfhXl$eJ}%`Eg6u5kdsd@kbqkZXXI_iW7u>kLJ-DQEkC-bbB!~rSlvBQb`=`ly
z*lyXC(oSS-e5iC{OuwPxXV>xwAi87#v%;|vdB%kejdx`F4-%?*CL<m{)9;|^bsnY-
zxkjlaAMNa$Qi2;Hs?jkj+AR2qhXPkh>D7*A^_d>`aV~248ksVBKr}w}AgpxsIZ!=#
z1cNNkM8i_r(IfGyR+xz>E^cMPHL-DCLY`E|Vseruwk^JU2Aah_DCu9Ry*T>3D7w~!
z<Bwr{$H|<{=j+~Yp$hHa9rV(r4%2+{Igxidtre5iW^mk;R_*0(m}aA|oLTi9)oqd5
z$vzO%s;NEql_o3x)Jaa!q0mC9`$&b`g-517eI<A|upTfjIr@IISo*L#00+vZ(di&{
zz>2yK^XOOb7y0*ZXmAXvI#%`E?vVv~z&4x(76@lq&2#gVNqxV(;MsfJ)*(s;O5W}r
zYey0~k2`B#Q$R8;i(KqR$~;Bi{?zz_jLQhSq~#F>rH{=;g_-;moCju-N*OqrY3YU7
zbljKv=|BGSgFo(Jjr4DW*5qa{1xE;NrdH2|>EG;At5kn5O#RnvI{j)xW#EN1TX5Bx
z)MsKm<acr$LkL^7`f~<Japfbyd1}uDIV=T;S~B$@n7G-Kvx%6KLC{TGlYe;J^joz-
z)#zc{gD=m(RRAXjVd;*GLA&sGKSk%6b}*S90j?tEPGN9ZVgHDiO-E{wn>}<G48~A(
z)bMGr;1u!reJ=SpcexRZ5qgWU`5ZVs!MA{XcLa?mHUIoG)YL%8VR6M`;&*AcbRQ#v
z&yLBWtq{|Vc{gl-PNfqCOI`CDHwbQ^TG(5mxS|(?<Vc*1Y_@Rv;&g6)sMKi-y*xYM
z_%pc^piHb2BBEzI%pJ1&Pl>Z7>!OVC)(Fb{uP#V{E)IExd2Ss|;Xj<>3PeUG@60?4
z<a57_V%{XUQF95%8W*U92K=XeMFU{;2O4yG9t1Eg|1DnSHCmjx&+JgF^U6^|IKoJ4
zh^&H<JUr_+WJ(g|jbeHHlz*!JgW{G!h4wwo1k0b{##^e!b$e5Z9W9R;y7CtgAvW!1
ztKCd@bazZ=66DC-6~mQJ{H7YCc~&b-7`oKA9_i)Kcmf&FGny%#W{HWn5{S{-bWx(h
zTgcdY+PQc7(l-*5WI5@D0Ax-UEIwadUh$P%-=3M<Nha$DUo{aph*NThZk6|I6~ku}
zYN|@h*l)_i*fj0?t4#n<7zBf3ke@#Kl@%loL0U>=%m7*#L#|XLBWHL`*RAa_njNrz
zpOqt<WLm6Ut54-15I`i7nvwVqu;bC^n4vFAI{s3h=Soj(1m}66+`Dm<lNLHrkJ?dW
z*)vmX;^*beRux!?4Iwju_x3j?bYCJ-VomFyt27j^BROew!{K4W5>F-)JNx9nTj%?&
z%@wI(M9REby$9MKxbpNbG^8jIJ&9_0o8_~7KoP46EFC}vw*7PS0GRZO$^rG@R7_O7
za^7&)W&0gfM*2RBHjQTqEi7=qHK5uP7dJ!2p}J0od_mM-G8p(L+_(<SEpJFiOu)~p
z<;^VXkxOT#!?mP!9Y33QGn7OQeY`*M2pOG5chfaLGvhR~Iu6}?q4N7Dw^K1;scg15
zruF?$fOw!<;*0x*$5G?c*p1GIFR5@k%t^9c7&*qLi8lOjQHrgvGOM!SYyBl3jGmV%
zHLQw}WrRhshJESO5|t!?Gw4*c7SOLY;7u>*+RyycdvQM4)Tr9zg;~D}5zR+lGxFZJ
zQ=Q;r#Jy01Zh8YHN$}X_Z-{EEp~3#fELgyh?)^H}4aDw1Ur6tL{u`AYUeIm;mws1z
z6j@8{&r`)*n<&ovk4D10YfAX4ux(!N$b@Oq+XMcP>f>v*VIa;pWrCElG$}(XQtmaS
zSk7qm5eycejfv+2DGFr~(c?lb(H8xWiH+nC?!PixpFs`0m#;(K_AyZ>9|#R)&ZN$m
zp!ECF+Zm0RD*;NwwXr6^G?|2#$ej!&lq$(jbJK}jwdLF5T%Px$iGA$vlh<6uziD`b
zkc3XTO2mt=mShg_%er#+g<fEquOVlYt&A3nDF-5-@LYKTbF!Str=3gV4TLOC-uS<C
z%3WP&Vbq1yG)w9uHC9df3lAB<;)z%=^xmwui^vILLbt{DcK*$A3RX9NA^RjJ%(HiS
z4QueW%n+O|RRB6C?yENO&?ocUrG7rTgEt;(f~=QBx!TFh>2kLuAEm#rKKudfk#YJD
zdFTx{G|T=GCMsVngXCs6a*BplP3x_2g;igBdZfriMnTZFCN`|_oraP=A&a5e0zair
zJ7y{zGVz{>B?eE)1Zxl-A4tWnqOe=t2$>fFM9NR7Eo~atLLN~aS&Pa!+x8)~bI=2f
zJz`kDl<oES#$BDv7aS|U4DU$jC;c!PHQQ=W$cq1~^&K@S(88P8%7b%pjAoavgUUk4
z`GPSdl*Vfzvs1r}=!S*nIFr#aG8L)s)@68aoIjkvyE;@qE#xz0<e$&#W40nfEun18
zw^8*C0#2qgFAB-Hcnd`P5+rt*)zlJ6<ybp9H)}NEu5r)#Rc|QyhKsClWxhod%!!Pw
z>WN?q_F#4X?!o{M+`9Ck<0hMzj?RwNt~x~~cI1y5h<4<JbxvozMOV`5&#Xk%TErmp
zkOJDklvqbXYg*vP-J;L=4B@-_gh@lfm{s*rkif{y0%JLDcT5HDfl}96$j@0t@Ryft
z&LpCL*<f0cL%mP@)AWY>DS}qvHp~FzN9SC+O!uhlG@238sSAh}-%b199d5`(x&2Ql
z6Hkxa1}c1ND+|&LLK_gzkF?&MVGybU458=x&rPh#4kZ)YW%#L`xdf?8wuDVmOLW1f
zq(Cl%oX@e7_*lv=^#fiNo^VJjl3Oley=~C;9~JC8U6d<woI;@e8xZO=o%vAw1R15G
z06t&Kk;DWSE+d@IJ+(bTDpUeFIrI-EPY);|*EO@y1gPt=3O;%UgBeZ+>bx&U2`0r~
zzgs1Y&E0>pT{wbec@Za?K1}a_4Kd+_J5KNVk=<g>!(}cAaf$!c{?3oJ)ur(Ud}z5*
zbE;bI&S<ATT*YUL)2rYCs0dErH3S-yh5E3k1s&tj4>!(MZQA7L^TQ-P^TdSa0k=d<
zT*83`R0*<v<<1b1>n*kyQHBM<U#Rs^VJ#ZQ=yQzwXB*$npKoav-V8iK`ez*eYIz#}
z6lk<^=iri;7~0CN)XSSJf<JlZEx&U5C_WW2w%y^j^EtTB+FEP+`d%pAa*3405Gt@B
z7n}>(dC01<3j<sI3fg_4&2Se*?7mD6jTZx~jT~M8(Rg|LQTDAfI3YFp^lJ0$AX596
zJE!XKr3cD}0rQxUOYghXWey1mu_k0{(|0uH{cpwy6jrxLTieDQ7f11qz?k@Zg4DJM
zR{Qs#CUv*B>T1-_?@ik9f<2)3;!KqDQd6?z(eJtA(kBo+JjD;K=d;PvY!Lpp3Cmw6
z8~c!gU=!>@Kiyp+Ir6WR<SievYj-aRSc$_qq9Rq$A0qu2ZoTd@Jr~$>Flgi2N>Ee-
z*)?5+Hug=F?e_c5b>R@X67o_bcBla>Q<{OtYr_3?<n}C2<Jh)nKA@G@h)N?nL`h2t
zN;wRbx0?h4fJ^on4X{m4xwud}_a->kagezMrnpZR!v%x{`_YaYNM(t)94qewc59UL
z&1?S6e9vnU-nm&X|E@CUGJCcEBSCw`c@iI0<YGP`%A54w(Z|A93mWXPmNCE#0n)I$
zz540oaFWw6X@>c1=p)|5hH+h5w(#tRV4$8Q+~M%DM^!%Hg@4Ie?WceWWB>R|B6%R5
zGw3<xjd)xFgX{1kI8$1upaf}(;$@eIX>yfOIMBN@XXUrSN+vc_JoR3R?3s!w6slBq
z(lKlZzVkor3_Fvf^WKpCfavhCuL733&0R~W!KEKs$0>ODP%<prCi#(d{il9Xx!+7%
z21>Sl&kcO1r{<6OuGZ2Men@ksJ<cP$FOd5p<h*lTh5TFpDffd<kJzQk=Uk79v3q0A
zkzAQC|Mm=G5?!pPbVRMK0j@)>oVj2-bScrxWW!r!@HF?)4`e^Hnb7=)Bsq_%`oK#+
z4>xF@JG<<FM%neu0@axja|5vmHp>Mc&akajztU%A7_FNCG>ACu-ogyEY!DKH-J2yt
zvw{m4Nn@!iZ(t=_A<?PD4GhU-S9D$(XCoIFQG|o5sC{ZowcS_z49Y@1WYvz}e@e1m
z1y+C$UPo;Y(}iZnqq&iz<nRy<Ms|f#7biSvx}zFNcnW$ksD|7J>IRXWd!p6G8;pbp
zVt1VCsY51b5E~}SV<GSvZ>wh<Tqnbd4><NqiNpQ!5|HLV;dTC6CI)s%c9|<<f|6&>
zC8~&+4X^DyU0C6i7;n@?eCL{vBmvNRyNjg}(#~maj8HmY#)GWPxfrB!Si=tR-0Z;X
znRZ~7$UBLkNZ}I)*-7<=_rPfLoV)*dUQw#4C(e9yFa$8&q3Xpn9WH$C?CI?lw|PG?
z%X7ETFg*nrzyeZJ24?-z9MP;o+A~m_`U?So4p(7h{}HSDMnSdbg5FI*+P9SaWV5XS
zsy)XpoSX>%L-}aT7bR)ZruNn;1d(m!d%l=ui;;*jZe!muNf)~`lBlSmqg(ysi%%rO
zvT@I58HjK_Pcu+4UKMKV<Y`{aL0>O6;-T`yKEDl~aP2uHf%iIw)*}h3eeS^b5$?G#
z-P3oY%na7jPHn~Q$J3(C#XZz1&~jI6g@`4#;mf^Gwhv;47)RrBXzfTp>yf1tKnOtz
znL(x3B%<_jaT(Zkg2M>lZ-e*i*?5eSKJcEYk;9K&OJV4Kv2v=+@DCF;^?FUGB<wQc
zV)rA)8p&rZ@zZZcFO*bn+COyS(y#)B8@O=BTi8?!;+)5ulo6BV4CJaUGSk1ONlwG-
zOp|5c3wu_vW$=M-CftSfTw*%TXC(iChwaC{1<CC8q3S>9ro??_6{>dDZ+DWCPJ{ue
zI<)x2e+lTvDy-%?@Qthwdx`$$$ntvi7yYy6%<B}AM3dld-_p&~v`_@8A_VNhvBunY
z_2+Ih3}}3*b<h4<OxVqZc(?$>?n68Td>jM^T4ZuM#wt$3x}~~;n+h`+VZRAF>Lzs~
z=4Yn&a_ofIvf!dbEGL|Fo=ERc&zQl>UgDP7{KD8AG+%<QqEd5o_Zkt?G0nn6u4QtE
z(aU|mq;2k42dj$XboN|$S7E|{DbyLqpuap|Rn0UZ&fh<6)+2kraLmWX{LOc!7*3C@
zk{1S6#A(WKaV<7=&_h$(G4;Jj^sr3egUNvV3LLFk2QQq4)8BiDt^Hi-J|B1YF#dIW
zUNAzp#BD9pz^du4Ndxg`yR(Z2gMF$R&+xl-Wz-FxD%PU86Et`Vt3@IjS#CUkQCg&Q
z_0D!G?&tUOD!I2VqN(WK0R;z?O%)xuMcjnD(LD7X14&AbVX5x6o_EmW+2>pWlGJ4@
z!@FIZUF*iexPcIulCag)Yt$>Qc~0@?z}lmqH)tz)qLS`9o*P}A5|PpHG?ixcVYW;4
zI@kOm-Gn<9So0J!3`6>9jQ%Xkn;jBTZ`W5-co_KZzw`8blwDL(Lt6{T4Gmqy#)Q*S
zE<=I71WtIIMy$qWP|*cFwgZ&OdcQ!@A>HC(U;JNKY%w33HxbK|!9+RulRVjD(0!AO
zR5|xduAEU@+(9A_xj;?}fp#NT4ux?5>_q>)`nw7(MKJ3ymouKsOp*NS<~6B~qh`$o
zW9hvVNK^DZUNbzPkRmUEK)6YLNL?N{WxW0Rs5pP(1w(WyRMsiAA_UuM4Tg42c5E9D
zAb60w&ULFUk7*`cuV264-p=AY9JRKbf>5L-nMh-Q%LG3Rg|w=e&iq^WO7{Nsh=uiA
zTClTuEphRIMr@4tZ=NgEsnS|(vL?K`s}u^9*~CFb|1_flW=9i2%WDbCNbfv=qe<?D
zFA4Z6w(17eF(;>89Xilm9FbdN$5%;#yiJbS8B^gMCb-&HwElE!rq}JwEHUbG+B=_+
z<&5}o);E;UIPw9RB!4bv4);O=IZSSEr`j?GXzJ9FGD`ae!c`g3skd0VIHbym$xEoc
z%GDyYG6#0~SRb&b_aQ3)y||_osgv5oU}jpsZnA5SthdSnZY$@JK>vP}#6`cIb}3XS
zn#OT!PK(zBiyjR-)U|G${~XeN76Gyh2a6ngLu8x>c3q2#S}D2z22szx@?V-6d*8BU
zFd=>>w`K4&src*f%NVM-(&FPkHNc!ueaqRMhLBG#E984*bsP^y;Y|tm7Sz<rJ6%)2
zs3~*da3KG$*6$o^CqYoN=)M0X*KO;$3pNl#^!S$byv63bOF1>RGE!f?i%#9VWg_$2
ze<6{al?7Y{8B<d+E=Jq?4S-pw=+qQeK`$WMA?2Tk{FT<vwqiha`1-Mz_*GD2ly7PK
zO})O@d<B5F>2@_;BCCJCm1)CV%o1>z>mN&6U_!qzr+7p<!|sl%=9#_BgI7h19<1M;
z2E6T$(kCq-{b{m5%<eG_wL-5L2e>$}{R>E@D0^<|TKcl>DmIY>E(k@#m-T!bA%ZQw
z{#yoW$;_`+$~?02(as;XUqm*rOrwJuLoZ1A9{xI44Pj9$I2zz#4e8*0z1MKZP%%4l
z1o!f<lmBwtpiUIed}P;CFZ>4uE?Re(8(eTt?vbF4l{7BDeZ6fP77ZJ#;d^)YRZw57
z8<(6=ixr5wm~Q)`?2k(ZTX5BTCi1=Q$z4V-1w2DNED`HboQc>#Z%^s0=d!KfMDxIS
z()vFI`=<&{K&Q_#ac0bjNLf<65}g;atCxL^CfFTO`#~D^|EZSYnuGd6j{mSfSw0Xp
zGJk0MK|vVU5dY(Nj|N$UFlzU|o={J3aMQZ{7<;)=bFf1luV?gs$9z-LN*t_ePAdA3
zvMD`JcNLv%7yv*56m~2VH9>EGh28SvY&+U6C<yj+70<SFTs22wwHOw;r3%j8#Uaj*
z9UuaHQF`70u!_7NYDvlKekyPe>ghMQsx1BaD~*5U@94)bE#BBWdOObHao<O}>%VAB
zkn6jb6#-0)ury2^C?EnkXzutiJ<c~%Hmu=lWb&)5l1?rgCPuxI$SVSF0;M(&Iz0=3
zzeqW4Wn$<TL0|_Fi`k(bCV;Rw_L7xvj@qDFXOFW~l<XQt2QsZf+S9j%Eb1krb!N>j
z6Y2v1ecF*;Oq0;$5KQfvrx|lfKFNJ12o;d;6pXO9{;5CPba&esoO?mPF&cZ%WJ#g$
zG}-hjvU3rJJ#gFJ{S`%nB$y_{;+mjTuE=yTefghcPpCx`5pY=wsA0X`VbQ#=EeVKu
z++C@&NqH+ZJ#sF6`xJ$Rj9=<lK?Wz$Vg`THyZNeNEeK#{lA{>;)t?&&t{+Fub(K+h
z38tW(;p@`Vw{@c=xClqjBwlrFZ@-ovFTg#QcyuZ%@EXi1tIvzDUu6IX;;)Ke%sKk_
z3z+c0=}8Lr-8!!Jt{fP*aFm&H>?1Ov-AQTVu<R9h=gSO62=b9Em$(aH#|1Z9CJW4p
zTwfYcx!!Uu?F9LAWLpZD)@IOzwduV567YK7Me`YKyHU-Xlhnj#`{~x$T3>>`_2TH7
zq@~rbz3$DX0!5L34$SbR&dDtcV5k8SF3#4%Vc`~&$|O$yQWhZZarTeUT&U7iFz3tE
z^G5I_$_dY%7vqJO><yC05#Rc)1EgrAT=QO-_($MIWLpU&tMK!glg|$7GG;w**XI&H
zHaH82%y71BXv-2Vg1AmP9i><8oc_!JwvVh5o4!+;j;cdfx~Xo{)l~O{e$Vykv06*G
zg+n?~^V4p4Wdmz6m!Dq9`Y*_vf6p7lFzRxS{rOL`+G@@&<gXusREb{cnF_xlr)8V?
zH;`XGUls=}8qsAtIRo+^v3aky^rI8if7|z2FKBzuEU;4*@jG$<&wW}^f@2uaO{DcA
z2Uh`4RlEHYZS|a2SEl>C6&=rFHNm-XwPiKSKx7Bk#dgIK<t%FOPC59Gfs_%u9Wd#@
z0M<OjVk!@);aI~cKVfJx9s%*w<PTCF<DQx*57CESwB5QMlLX(Q$L;Fi9&dO7d;0+*
zv14G7hIZyc?Gp(WeUKxj7kPtggs$rZ`VT0;2+^oL-d2qu$(^$t?)eeL?$GRkE|5oR
z-Snf#0ore!FZM(V<amWVJjgKDq03<A5e@pYp~k2qDqN~<7;iS5^c(a)O7kKyD6hOV
zK(>x*^uD-@;GROz*6JVqk!iKq4t>*A9rt&20hukpgJWP!ZLFZ+x|?X(^zoY%-iH_R
z`l7?YXFAUNP6pzX^rs;UuU2>g*t2?WXj#8O(Rj^2LA_Ek@(`P+HWf)qaMWI$ZrtTp
zCz;6&S92oeM;3hjtoMgSKR?csw*M2pS&|WMAoN9?`*7!3(q7Apyy)5zh_Pa{=r}oo
zRnU;zGlGSZ=J}wk3tiDb;!jn$Z58n0lePwM7yMgw+aKWVFSfd$3q%hK2Y1TV*6_<e
z{rbwqtMyXXn?ia_QC+0GCnE2}9}VB77WWr?jsgecPpb2^Hs&&(0wVq98Z$Pk?sSNB
zjQiWQ&aD^4x7GLeg);|EH@|yIDu6xEs%dBo4(i~Uw{(FShSNsy9DT1-HK#lmf;p-%
zsqSU5X6V#g)$4A<Z$+YzD8q{+GAC~kOWu$jE?~-)LAL~)$*5V(tIEX)V^g5Z%hfZQ
z>$&UmP)TJj$nlhK3Syr0(o{emmPyZw-(&3Ynht<Z%w5PGn{d$W6I*%KfTi-6{WcVs
z$H*m$oOoj>cS$c$q%g{gNd=q;L54_lYFOhJX54#(rvseq!P1EaFgU73=@mNz*EOGQ
z`d(VU1Lc!2!KmcTwf~A!zGB1fb85H~PnmX{79FdR1<SZ|T(VWs;0BtIGYKygyppnm
z*D(+H%&7l!oO3#6<z!Gas_Jw47unDNN48wz=l=_|5PFoZADud@8r--qw`@ym<;*6H
zeIZJbLqIM2vL^ZjQAO0AJp`K<nU{sl0u@nCYe^O%%rl_{lL!fae9WpJZLl!Oc$d~I
zg|q<A!+{o`pViPBw<J!f<HSTn#!1O6BS3!C_}KEx@M&tQS-WugMyxn)Ae&j?hxHtF
zZpB=*=#Iz$ShyRboWn#P{&Z20%WN!MSq7#sX4dyBuj4Bn+ezrgIPD@&BS5=cR(=?m
zU%|FVI`!zj`8b<cxICS5$Lwux4pDo+Rew3cKW+wXNkAW*(}q0R&DgOs+<s|`Gan9l
zCvoE$?V$M*Wl7@w-#)1O`+tDb+VW7-qX|ffDni`LI=5oLW>hFyI0PDW0NgfCJY)Am
zk}OM9U)M~MTsRf4E=J%M)4LdnZKkH#=dSc#^Sp!Z4&8v?{K=Y;);BjTMzQ)rbLrSE
zNPYWT<%WXL{OUx5I7ro-6DU%-u$+1GO%Gghd_4y4SfI9@AMGuYePOGA(D2b+DuR8D
zGbmZJqHffH{yT-CN)6GdmCPm3BzfpDx~96A7h99b4FLx?15PI>R<C|Gk66iUnkB(!
zx(wnAqsP<>m0|}4WdDz}w~VSPYWsH;kWT6DQo3OyC~QhfLTT7Y!`^gvcStuVts<Qw
zT?Qc`Y`RMtr0Yzc_kI3job&07bH?zC3&FM4UTbpCIq%<f#bl^F1!V+?avuTW8$3s_
z;3bIRZ^8)4%4K^mSY(2b@0l%n;#(C@go`<Jst$dY=>l^}L#gt6ogW+eNM$eR{?uW@
ze3|;41rTxK)@(m%sIzEr-}UaxS~s-|g2V^$P6Y2=*@-Z;H>)_sS~i36_ht+cT$3b)
zl&;2{paxrn&WzlMoUp3CPB}zT*v_c{FW^_~>oWk%j~jPw4q0_Rc!F#@2%Aw!pzBth
ztcdBpOyDuPB|a$Gnn0B3nZ=zz6%_O%mW^^GhZ~CULvcEO#OxWM@fp)7iknV}f#!B7
znxYqJz5QS@W-#7pMNj@45au-8Oi@fe-5g<d{i&2|K-M&@*!mKo0pbcdzO3p=-?P8%
z#>B_EB-W$93LYm-CgWJNdW_mt2qc202%WStIGDexVI3-s%h>Xgqj6Fa?FbK=eG62}
z@n-#qmjq96Yd!e!KQNrPmP{w(_h_F3K@BG&5cAL&k<TY<WKp0&GyRX4-5{kr81GQ~
z=Oio0I*N8aScXz3LMCErE9YFI>DgY|o9LiA1joh9e4d;LdTu~T=&Kw0`prd_jqNn&
zw<xT_0&Px|9wn!-YMJ_f&2ph?JjxOuNdWkUu5!)i(fT)zXjoxLQAd3Wq54y#?&nB7
zz&IhJ7+lg;3aW8r86siNbSpj(+tgJWQ9|&5_H?>o+d*pjQwy!4SD$X5;NWWNu_sbd
zO6a3Ww}opGzImNcI|#@Qv@Ue&dX9{_DqGlFj+!49{FjxS+BtCT6$krn+QepLtq<Dt
z@%KJ&wU=S6-ZoYmb;?JDKq5I<C%FDWW^1_~CDnoJAKxq)WHKj$OA3c>){e6U;Wa6T
zylQ*(M4vlK<oZb><Z7A}UZlJ-(Q_Oo+SxGYo{t|orA!-|rJO(~r~+VEGO0f~YB&-U
zeq!T?QwmUg9?!d9sAXJzX|?nwxd2lH<|;>FY+DOjEF`_7!?K<d@rbDwye8RKm}Gsj
zv9#+{Ric>{ybChkS$~QYaQ~p>pUPn<-Iz6VYGUmfcucTOMr1vgU_@`sG22O@QfSFh
z6c5I`8W_7fGN<TCJ~2~4d;U#h`SF}i@IyYEfREfmr<>1}=PKLo325(N+usd4`<ff1
zq~ml~{2OiLLa7>F?;Qz28%N8R(&ehE4;;kn+dEnV2U&_;xRb2S-kDpq>36af7kv&Q
z@08ydE$rvhD5~Fx>Ja;MwP<Hjb<V*#tC{dxN!A!N@*%7D(#MUV(JaWwoZotE&z^9^
z!B+Ib_2>$hgJs{pnw$92i^VCQV((|by8KQNr{Fmls~*EicG9#I)~FXw!*rPivaLoS
zM8cyGP3ndp&QwrGqGK-9?jm%F(mI?_>3_ew*Nc+X$!2~Axw4&jjzmFRmyu{hcbc>k
z?J>t3Q`P1^8m0zMKkc?te4E;dB879Inim!q_fYu>YY6p3fN0l^`g9bgMut#Q-z}J@
z^+?%WD<O&<$KE5+8=5CerRPk$LY1}aG@iV*gMhUV(%gTBIwpYWHxT~!fAMw$@R}b(
z%U0veMb5{bza7DH9W~<L=7?!OXFYi;6BdXFNxsBt;3B3dXi}F_+!jNy#K%Ka2cw`_
z-YqPhY4ZO|wTl)42#(c579t6%A*-yZxt$m!n@lkMQox%JrG?Q3^l_0dn@4cSdmf%)
z|Dt83eLDP`6<bz;tKLAjA!@@!zEQcM|9uG=t#yk2Iq!t%sxfHcQNR~aRDW!@mqg>F
zY?ktL`hebM!Vrw0gobM>pI&~_V@AoT6N-E~O(6uU#Ek}=yKjHG+P#S2&`i(-$B0?S
zp^m{^Jsg>@k$~j0nXU@BZxqsGbEVKUNaV}{Y`SsdBv?1rJEmXLlr_>Z(E^7bX5&*{
zB!6*xTG|%(_|ae8A`wM^_tX4QUyyYw*F&t3-ygE4mm#<-?xt06&SLrZ*ci%U^YSl=
z5`B$5YX}OQ<&sj5Q9xE34}?++`94ILeIQe}#n?$JGj;VDKiyR1#QJU~qh37T6+omR
zeVMT`7=%IRFa+5@s=9mEqtCH!mbMyMXgDHILQ2o_AMor>V@@e1OO!A<wP{Kkx|1Ar
z?V-BF1+JjYXtm}1w_jbtq;ta`$&Asc;nlqrK@0|DMAo}NSE)7Vy?uhi%TrrutH+f^
zZs<7rE>$I~OuMSOftK0MUjJk3cL1+sB^?Gy7$z(|I{r{uIToVV%HBEJWN<&~uns*5
zAsUu`(=ikc5X47GKI<iooy@YGnLNwCdEwIO7dDxUPaKrIp6C838?E*{qxqNAeLi!0
zG9y!6s43?wD|T1ZgCm3DQ?V-UzpTj<HRb|TejzQ~wm#$d&ewYl`WlEU84icG3s1n3
zp2YC`t6X9?sW5oCt&SPmJN~9R%~C0z<tGsW^KjVzl%#qPK=cy52fn^@(XU<)c1TY{
z@06E*=I;PJoMNE<_CVV>x2XO5$wX8(e8-f|n}eP{_YoM#vzv4(X@lv{H{JxTkJfJ!
zgV@?`mz=boLnH^f-4~*^{@>Bxn(R1r*L*;W$9G>dT>u-2we>B>)8PgH)CUpl-0DQ^
zNPrJlPwV@!L_xt?D?Kt*Uid?kkC5pIfX1klA$&~OT^vTX_XzQH*^i|4RlJ%3uAZ)v
zNt^08F*_w*n{J$7uEH(<LYR`sm&{EaT1V)Jgw#6~kETKQBGpnLrVoQ6<M^u*hw#+9
z8DDPMyC4#L!Z=Q^#Ca7oc>p##*F#EjQ+*-sU*(-}2!=XN#UNg0+Nxp!%}C#^5^5kA
zQm5-RMK(3#bz|<k$IX4lRN2ut{t+a)T&siDamLJeZh~)EB2BKh!oV_Bve?g!ckc0v
zz{qAocf|#WdJ}2MoQji~B|;Mq7u_Xf<W2OttK|Ym<fuwHH2@q}@-TL+slRI+^`4IF
zX5<vit+ptKMbBP8W4{huHw3#TphZ>8oyimG8~IJ?3K~TW0}-1d<4S!X11BVdiYqC6
zwUf07+D^BeimnTe8+zwgI74YIb-3!$rPq&JR84Nlyqrz+YDYGg>7(|wVX+%v>@t}<
z5hRqd#XJb~+nvr!SW|rRA?x%H4M}r(u}R#K|6=9ouRo=rBLRuiNdXv?SR%prD(@4g
zS&j(pcxJ8G^L^#t>$Gv@n4Fu!?#KZ+I!vb%y?m<<_)S0u^c4tdT%yS!QpDF%#947p
zU#$-aeMrMN>>C^iSuWlvASGf7m0YRIDYyd@{`fd{MMcDkjK(8s#ROfj4abQYeVbgs
z31xZWm!sYMu3@7^DG{2E86|_EI{k#|1W@C04ASnIcKS2qoODB4n!9fvtH8g4{N#VA
zwDLdXr?E7P8mp7-Dvc=QyBlVme>uJ|k<e;;S{cdLCZN*Y7%78OJLkn#0xsQn{#T;*
zOX1yZAO^sLc83rVjr5gd@6j4pN`6pFRG~m60KhJ(>^1{iRH+!1lyUTI_$WzQfg(a9
zD@#Q6y4)FPaQYVw+h&($MUERghIKpw{~A@TJuDROIwrX^<tm?%-?;F9(Y5IG>I&+8
z1Vmt*4gG+nRVWe!)?)N*>WjVy<2@wvi)>~=W_Xe449nsrfy<uNzOwB!vm=a+6ourP
z1Si_Lu3ePDeqH{)*1xA`zhl@`=FVk$QIsC3qnQF0*df;k-QAhang)-@mx<Q8>0Pdo
z@wtgVPuTxzDT(9s^ZdEycWey7vGeibRY=0P$7Wwm=tWG|m%V#nOGi+~@rYg_O_BQ7
zVAX2scp8amw|Y%l&NvzC{`ah3-@J|oKYi|tS{6lB)tq~v1h%ip(Xd3HczN|uCpdnn
zVBXD={S!XTC~?6hhXmW-wH=;n*3$~XB$-TLKgTi`V)5(=@}3co<dd>d9Q}mZz@&;<
z4U#0jnlGaf<L!)OJ0buyS@0~^^VxVVS#O(|&cLd?uEjHl@ky3v*q=f8l+mi-^ZlhA
zA4qXRWKv>M*St~~8cg$JUPAfpurt0gTz`c;O^ak9BDig6NS}0O@zPYilBsj3?zKW$
z`dzK|E#S@tg<6cYyn9*l<(V5RXg1;cr(*Q=p3YykWigsIzP>D5PQ-7?r)op?qF{+x
ztiPe?a~z{=ykKSfZOJ)hTN$KtJs9(?Wj`R3V>9bvGIyQoG3&Q|z>nnY2AoB*&Tv5%
znw$o<kDDY$RS!cz%Gc$QfuVlYD67b)hicIrS-Izgaf{Hg&I*{zu)+vumY>s5v7BY{
zd!DGWH-=AY0M3TPGPC|=X>1%Zw-V$4%E0Hj6!j<)T{D;K;H93jrqE<pM-F8{`BvWo
zG#dEXt!j^l@&T_X0sHRm-#ON@kDXHe?)Dn}4HB33k`vgT)zO%i9UktZy+wS@eAeQn
zDltT>-tvKrh(As@!5_4_kRpoISrhYFkejH=Q3Kz4sr)#9&A}>F2y6PUzcViYz+33z
zY};?=_ys;x3_5s4i@j6C@PVokjtF@U`H3@$0|B*^<c?V8>#(l$(@Him-MrvMvLUv_
zH()2XiL!=^R>Kuo6s95{7y8ijozCnNPSZX0ZRyI{9$G~mH7U03jYd&RS|H5L2>g=h
zTzVR{u>n*|8pxr2wOU0>4UPgoqWa#vQ7Jp(3`rSK!GJYY^jb{lN7XK=PvxWRq6{>O
zaYRNXMUSqSykBT|k9r%Kt8YF(o85jj_IYya1H;znTEIhT!B!u7Gaxr}J@l(DAUiz)
zPzlr^5$W<$iSxzG^{Zhi-aVk(;th}K)ijV#dy?R4Lp9R3+;4ccc`DW^-xtZC;@YB|
z_co5|EjmU@%_{{?yq6T#`L0J7ozG7QT~;+}L(W(#`8Q(1iQpO;eFGuMqVt}~e74W4
zd3ryOw*3i=#~bMzWUX%ju+n?gk7w<9nN}I;3CtvCAJ~0{{%tpY1{1!p7aP?`kdLt%
z1O^{!32Jl#NngsA&;R%}>QOMlTuN)}-X13@SK|MVT4Wug{X%W5cH=)0tV{!k81-Sm
zhn*(3c}|zvjcH<=(4|QOFhhkD7Rry|r8xqEDq3JB82OH|{9N*5>*FSNLR(CMmiN5L
z42mt9cU$tF>N))V<@FQ7L4)<hg<P%Pxrn{C4ZB^bmbCGZyRHesm$NhGMYEa_&LGEL
zrC~$+ORO{8^QC^j5J7b~hb+LhpSRhnUq49?Cx!F47N}5LxsR=9JJ5x1aW<u$o?wlw
z5=d74tJN9<Kxf6s5$AMFjwj#_Smg>Y{M$Tie%}o%+IMG=7L<+HqapjXto_r^(>bvt
zaVZv@*Vw-;m*)r6cSIM10YW!d0TP?zCZr1{LeX#&S0*lv&S`5Kcmzp(-6lK-#w?)n
z1Ty#_I1+R6+28{~UPT<i4#uxC;h_>DH7v3b)tFZS%<fhfNZ<)DUf%c<?ERO%TJ$e|
zX3#HjtE8Md6DrV;6pVAPDr*4|M&>T{yA6AnkJ{yl>nw9hH!G<+Mr(eY#ox9_Kty$8
ziwIG8?z8$`<3k9GHoXo(68kJedGl-A!(U+cW1~l-cEfDdz%JU}#1$5Tp*2jnE0(2)
zsE-&N$c#%aXo#h%_xZ2shiMb`G$7nwxD4Yi46|_k^LSwhA#8`76b72WV`WFR9t8>`
zGhRfSq-e_i2^~riI_wC_CxchSfw%?Hf(g#w`T&-|PJJI}j&6T#&Mlu(9Y1H7zP$$k
z=$GZkDWB*tP3-J5M=Q4?gfD&QT_r#})<18VgK-KxN&yxw$6}uT4O}Vt#K<p2U*lLh
z`v$8(uXIFj%m0Xqm}1+8B+_GO6$J0_(NS7LIWOdc!dx(L#a^9%LEzFfDKBdh%Y87u
zCLh5sP$?$sYyd6ZjR4km*uBB(rad6QjzT_Z`Ub~$b?%G{i=C4Y0rE6V@S3^KaOzL6
zUpVMRlF)ot8*!pCmJ#{CGe$oE3e=pv=(drjI$T;GU|LM$xEr8V^3#=ho8#cFf0Ghy
zpfIgM(r@1`Hu)COI8_xYB;Ze`Cr<+}ma_c#itOB&rt8&P@vuKS2?keXnRp`B#*}Ru
zF#h!*)P%<YpDj89)q6{1qoL0(cx_`i2unkm)vOP#rO6W9z-jLW<M0YK&iN~}Ab6lP
zWjdcvS1AfS-8GDRDEpt2gr@AA<z4uT>6xB6X`xVGSzu&`R2Z4?0h;Tl^Qrpx8kY(?
zasJW8+@2Gn7ILHhkDUtQO05p+2t1_0jtyr*F>J?9`AxO_Y9<MnjXGQKpI>jUY+iM;
z1%5dEPTWIe*Q$Z5QYJF!`mgSlAfYHR8_?Q}*Hmo1aUhv%i}d}s52A=(LdPYP(#FOf
zWyT-IXe2$!mm2kNhjFQmb2t?44PS~Lnq4na7H=}V1?ULY)L4Ly<Zm#?--MXJ(T^Iu
zZ{jwX1Rk)(1Yxd0GQ~RI2li9+4sMpBnJ;!!2jy}nAm_3zyYW2mAl6%c_3w$U6nyDU
zL-90^(@O3@s293-!0t}3Q>6+(^W9&)^?InL9!&U}t%ZKG>?Fe@g|l}2IE;l_>fq6e
za2eiVGAqBdDGA!7(hw<;RG0w|*7g#!JjqY-i}!Rpc9V1XcnYVx5Q6iBt;zks9!j+n
z&<mQLWj|+6>|e|O%*WUotdBu@&q(9EPfW^f6E>6RY<iz31?Aj}*@_w{9H%GA$aXC7
zo{h@2jZQ`K^-u^NmNgc3%a=1w*O>DfdmctuuEco=qoUJTc%Mj!5;{hp-?|Lr4;NMl
z$IVJlyZwwbY~D`kWsDL*$8e9L03VPpLl*ojMCk2cg<$_#Rgq5K_5dlc=}|KJ4u7?h
zWOpr$6Ty#hX}@Z<`uXIoIXc}5eB`^4cGZZK>*jA{Kdta%Fr-DUCo+3rqYV^1uPVnq
z*XeKp?gRBqJF*_>ou)jSJicSfvBLh$P6rEaF3a6kBG<hB4Degvg|9l%N2|Nu?kc5#
zm+Ior38(ZtD`2uI{GCrOg~DE~?lu_cCG<ykSJcaw4ryT#W)zA4z$Lu-E=w}XSi9*k
zlIIB9!UaaCxzXYUTxLD4uIpuGX_B*$GHd2Rhe?Dh;C00g*X=)Lo#1!`Tf?xO_hgb`
z9{8>vg57}4pq=2-b%+N#Txq@}z2vT!sm1B*lS#ABm?DLfz-k;>6Pjvv<5XIa<h5Jt
zCJn#gjdN{5jti`W+=7p3b(jMS+qQq1@VD%FSKZqnr(7yOu4ga6rC+yij<?(~38+s;
z2$iXFKkS_?ck5jM&Du46Jx^NbUr~o=d~$v8UEgt_snDf2g5NaAnpW9`<n^nIlG>Am
z0rPb%Y%aVmDA6PvH&|X&rJ;NR)*GKfcj3R8mC8+9iqtZlUal)omnLd-rV)p_QeFt?
znJ#t&nQqc6WLjEce}F}^cnwS}MW~BD^T8L)cDPk?=?pvW*5;CV<!!=TaBp_i*ui0x
zXICidv@ax!RT}^adwonn(aVkw0#ltK=FtNS=&J6kOv+Npy+{s>t!b{1aru(kOJ3|N
zC6BksDvqX<1dVf~2wXU&BS<*juyVdRP41x@S(@3`rI|=N_7r;K;Wz^!eu9*IUKd+T
zsCyRK>#etshstWfIEdT31zK$afFzh*C7tIC{TtbVUu_@NF<1`{8E7?WuI%=qBCY4`
zWsfsA5z(}y_fqJ|m#FsIFtG*lDi#-@;tJ1YM?j@ARzu5_YvDw>Eb3OKOZs%Ia=Sdg
z1nvd25qA2iS}3EGn>)8LG68w-UcAgv<V!HBMZBrw52O&Z)T!NVARZ($*dhLx*b>%2
z4G@$2UCc;ATU?><w#WwHK5cl?aeSrdH_a-0p5N=TEBlu~eAN@UsY)L)aesA|`iAg6
znJ#Bjw=8?~zhYMq7-{9p8h(t6K`%M?Z%c3lM9@<^93*LQ9}Ek3E9{WCXYDYnN>;r}
z{~`ERZ!!)Gw_H`}Q;b{rEJt2a>}910p{Z4llO;t|>#?#&80w?6Q;>31f7IY*214?;
zRAA*{fgd!i7;FnxW4Tv(3Lbd;bAf=>_)oY*B1{PAY~(k7rw=;3WZ@~GTg9t(&b`;`
zI$Hfd7S;m4#7jx1K5_13SV7wg0=$FkrXSko@;M^?;@Jsc?5#?3TnR&kA>)ej7iD*U
zF>PMueHjqo@!5YSTs2l4oLj+HsIb3qsViMzH)`ePey=v}#?K~;jdUQ#ZF;s{(PC3d
zvGvK1Q_2Q<^;Y<@Gvti;2EDwx&j>6!T7?DUnub4@ElaeNX%b%ceqApEtjWysd(6`u
zxD<|>OKCQoR8Nq$1MJe`oO~1e?tJqclAbdYpM(Vdb7M!^bnyXb`T|&zEyJc9yVXN&
z#^g=<t0&6tUII<<&7&(gkXkEuS(p}`aSj5-DL*kg9WBMSD1^cxMas|p*FVp9-g3Z+
zt;9S{Uc2M{Pmv7Xvg2HJZp>;j9>7|AGJ%11dOjTKCj!LK5k6id=FA_0zx2&kpDPE|
z=5kUwQU?h5UB&N^2-g1Zy{x92UhE`()IKsmWjAXE%#YrOtuNca>8wYw=VM*g#1rpK
z=$d^5Gyj;LGY-W4DUsk3&^91ob)$vxivdCJAwRhc&mSG$#D)Jy1zlgc#hz#P)y;tG
zJi#OzboCIY6KftR-00Yl8T!Of3YZt%i#1WM3~evZ@=Zb1ZfW<>IxZBu0i`2}w$&S9
z{YEeB2gaRK&Tgg3*<L>XN<aOm4R*+%doT1Szv|N!|0~*inY82%+Q%3ptK5qq-5E|_
zw7h?GtO4Tz#UX3f_ANI&fFZ$6DxKaU2ri%D;GKD>-q~IV*mpHUeb^|+cSK=Pq338+
zs$Zmee=iGX$cH8JARcWOJ66Mj%6qN6f0&wI`j_2l`T{F5?A=#b50M{(uhonHyNo<V
z=3Xid{qI3i_{$HkuE4ka_aL@N@vAcM9BKUb;G=9!j{DpGe+~w#a+2IX7ymuTsa(Yw
zUBChU{9)ZQDTiGhoY8YelmGY2|M@}x&+pObEFr<ZA71{?4}xi7hk=iR_`e6Wt2o1C
z{(F}92mfCv;Tt#{<}6BLDNN<-%)*=Rnm12BS{@p7T?rnouiy-(@T)Fa#pj9jbNPD}
zH!Vc$x7nu6OZL+U`umKfjCqhJ0wH+mklSuaW;KrSPSee-RnM7^sUB=tm3d&cbZ+<Q
zpjmA;V#MMud%^Maqr-UH>hSk4QuQpoAl3S-A)@2~)n7YOwk9Wvhu4!c>mMP!B4h0P
zrt%mIGxHI`$LnJrnK){a$XIHl!Ih%~Y>_ArdLcoBSjV3Q<bx3InS3J>itnplmWN{w
z8EbpD7Dv;$TBwI>6q2z<)_HMzjjKX}QjVdjW+hp6Y2U1k4My5p6Rowcyxm7a)3zb~
zvktxPFsrd_wv4S9I$_>cQs!~Ls%1(+55wBldKdTod8>FQ1iyJ!^M;XCzJ$W=FZzA1
zOtR|SK(l?jlYDU2${N>`5cLIXHs&sQb%9!Y|C*JvwyW4?|D(-CU4`r$s|8CspNn0?
z%fAKhgH;QJe~NU^(n};a4vxB%W;PSF&N(eOW5xRIH77JgR;C7V)D8{#TM51ykITL(
z^BVID2pKIqJsEEvOnf9?^3C}62w`~{)tG7Zd0wy8l#Cwo)$RQvcEhB7o_V}p&Foa;
z;j62oUh4Ids2|J9yPy3zgYWcj|49n+h=rGM@-Nt76sT$qstya6<bH$^dkPXNeele3
zPWN)@{k&d6Y?YC{teMK>knIz!MwA*bkTlBVaM9wBQCxdwJ++`jfF{~|OVf|^TK|Y1
z|EXE1#P3l&iwdc57|PaQ#Cj%YknXT@Aa8)TsSBF>b=PC}VBKs>GG3q2f!+#~X|$P`
z^{nc_8To{4SUi?|sj`W^WNGPXYOeO_$*hc^NZrQ~%h_`q%Zb5+tn_KOPxp8B;f`Hz
zA@|m4lk@x37xdFyPfBMq&0xi_GT3&<(Q~*kB+aRoJWG$`bSl9T_p>D-yt6E5ADhyz
zs(Ib1CBUGyKZA2j|L-j3@w|nxE$ndO-RYm{OjSX0E{3-fj1&(ruTqsHIvWHiSm|lR
z4AJR=Q0=S5pCZ$wrAp1br5t}>VKD!dSLRfl+0M$ENjJ<EDX>XuUT<Cxs+iWH)p~Fz
z=PFW{d8jq`FY4bH)feqeza=euXKK<Gfsa*7Er(>HL0Cg(d^%qQ9YEsVr8_hqxD4lx
zz*!Q7Eatx*9~G`=@$C<J?3(geZ)x*lDv&S_3}8@1wV9Hs+qg%ovP&V~kdMNI3jCh6
zz~IOAxzC0N5K#iiahZD0-65Ox3<%<1Lm|x0qSyI&+T~3A6lrexERXf8A$+{;TBc!f
ziWMh&J65cuM0}mUk@(HkCy1?s3HW47gaY=@<$QTV)CDv<8=MA(7aVwOeniK$^ZO#k
zO8=EkO0xS|`V|CocO6Bn3z>LE-(hu!llz5c(nYgbwQtzN<GqCu4kcXPY#0y!uzl~+
z<93*vlz3WSujMx@zjBTJU4PW`^yE{sjc+zyt7|Rye4HAOSD&GveJ~|4`Kih2`^?_&
z0(Fh$8c`xaY$Gq~>VHdcmULR`Da~xBOYVFQ-A$Ftme#uZWOa0CQlxRF(1USncR5&T
zd1@l!Ig+(!=$@WA;1o+ija^g91K)yhnqe;xBNtZJ&&F*JmQhLJy9@+-#;XmJyH*f>
zh$YnJAwoG<kH_m6rX7e1*v}Yj<CKpUgr#QF(zUqh@T^R}OqR=;H@4Zk3ys`lm;O|!
z+>Z$o?3qWQ#<3Hxk!Sr$44=>wohckQ4k+^*VY^BwYN)|XDE?vi{`7VGs;SABt(P&^
z^PbZ^I(E|AXj%s*m2+N>$5xHlBBlEi7SC1*KKtlT64~-pm>U@E%FTGo>`VU6d4iZP
zEzWSiMy#`*)xLa=*=X}m$VGjdO77uOS*Lp+qTcn3ysZP7=~jaLhmoDn^`~8w7jr_@
z7hgpN{cE}vwOl>bO=zZJwXLlPvShk44(_~nj}aJZyq{gTMQlxi{fe!h;z;~~niQYD
zc0H=9Ot;T7Ba671bv{Nm4{rU@Bd|~onW=*(lbU=9{vrp!Uo9ozl|Fce04_+{&sN%1
z^)&)~<!O@Vkyz{A2HI2lV`LP+qRztP?W_`q{Br#CC4Li4yvk}!&2s%s)kW)()y$p~
z6wSYH+V=?E#vyp^g#zvR68k(w^b1YQ$RC`XmU$CT@hnXCrKa_dK?mhbPpTDp-M9%_
zNT6Mw{k{rI#H6i51<dYi)1UcKRSCx}4&=PorjTHk5;yz2HJ!o4j!JsFZ+d3^`$0PD
zhxNP-R>bh)UAtyI2~bBCvS)6^<iU+TZ5Bf|&=wZp%C;tMibJQQT`Vmnr>xIDUX!!R
zplM~TXB2n(q)SWjl31_s)Sb%aQv2CzF`$Q-=^vUOS0~_Hn?K3dr3cQJQsV)Oy%mJb
zk2e%Hb!8XT8v<R{Sd$IZN+lVrmlMhBS)7vffx`#^GqxeVNgP4_c`qJETjK09J1W6H
z1Ab*rZyu#vlU&)7uJQ!!N?OXE;oGM?t%tdwV}hP<g7im27s&P0?&O{7d5x>L2J^w(
z6e?0e!!&{w7ECY3P6xXNZfNQ&zjj6Vbt;Rh3ux)HOA9AY2aH(hQ@?d|bg2W+%rP?9
zW>8%e69c0$vy{dZiCG4F8k4#{C8yPm@@s^cJ<R5({1bRNYwlc|CXNVuS4ok=?$b6|
z?mtsciZ!lYR(&81TC}Z-w5<m|X2vCz(vpwsbZ*Yg@A1_gJ-L%Lrm-5(ji>MHizi4>
z+>lzyI%Hg>*U9zd=^w$ARdbRa0ZV<2@06hBVRJG{wa5pHIUz4;R%-Cy(c$1&<Wl`4
z^ho2Vr;Mw2e=V$Wm9+EOH9HBvp7%$1j@y|2?^tawaK5OT+Nx%38D>MRmFUC4aJWRJ
zpPQj~;hZM_FFGN?e$lX#7gLj_T3og!Nq(tS&sc?RO=uBiZmrKE<g=RfRR6z}IRB68
ztca_#<i7HXF3?VI3aAd=&A+>BEb#pQUGo{lUCiy@!HjQ5o~{>4-^s!E;rI35*1+k>
zM*H<2-;2G<tmC1U3CWAx{pc%>)DO4Hhf&xKR6(yG2OWZg?-={D{BL*!sO>#N#V=Ns
z2m1r=BANpJQ;RBD)mMrxT!&40ZzqF29jUipT`!`x=5O%9UiVb){K6_w=eSB9t`ZC|
zM&ytl<(B4=pJ_;dl2(4F;R*|EADcWH5)>QZ2CddH>t%(*#Ao|l|2env)B2To_D53d
zCcD>zwx3(=hs0K>oC1m^#{*IUTgU{!MdoOci~Wq(ep}a&Ucb~+FfVur)>_NJMtb&)
zxcaY-ppC5jw%f1gZ9{?g(}E#u6R>qHIM#mWJF)<8UaMaANjZ9TmpTAx_nm}Sb&}HV
zIELWyi0<G1odjPJ0kg6|EwyG2zEK?r1wxP@%fNo?TVb4hLnXPf?++l{B|HA&@XWYO
zdwfCa*2hz261~3<G+_&2=Z*y2Y^Lg{=1b+1`^JGere}s;-6s0+u7k9x|H<}3<i5^m
z;f<Mq$=7Xv0QwrE-Fb(oQ`*;<e$~02xwij>FX7o~lU~z-gv&J+dqxy7dnP6OAU5v5
zLwtFhxvz7UX*@S2=X$w8pu%mxK5<wNL>&@n^)VqgT(JtaySAvkIc3i>`F2y$6d-(Y
z!Rv3+7c0&$$rk@bJ%4*32qO=1d)b)Kd$XN8wBGK2J#F`4fl_LD)7`%$Q#xa$P}(dW
za_i|b=aw&OmFw0Zewkr5prJsID53SpX+9{3889+wn~}T<*%)wGxO;x>Tl{QVy6<?p
z#$p{%g@JzUCKbN{xQNfP0GgDe?hlw?k9@ls;_V&Y+yT3dQ3P-Nt@FGuTh{xi+hJ-F
z7pw0>ZdEvj9K$D#Tlh#2?RvMyP5bHlZT+~k?#>v6yQ6E;VBL*CRNa5$BYysZo;CDJ
z!l|#=^3UsqXv6J*j9hn#k?WQGyT=P!c(=C+0qFgF%#vhh$TvnOaUJ2j#0y&cu4H#7
ztyXRH*p)8B&v^eqzy=2RxF06oNJmUI*-hqWtBU4Y?gcCcIi_9yBUZ;f&9pHNxQRQs
zff_A%uBaY=XHLl4t~4<R58%M70lguqD;k$y+>KelWN)6a{-i&CU+?JMl>3a0<#^f3
z0K1fv?;>+P(b;GAx&U}MZr3sIH|vVF^{n3wi#5M^39kD1&7`Q23@3II8@@D#cbV(o
zj@L66?v@fvQ|^|PdH8|Y;Kx9w%NwMJz-DS&X)Rz`F;kE*4*`NyZ;t;ZrR8W5s*GcN
zx`|Vm*ZGpdTLp~D+w=bQ4pahx{FX(O<IG_d9GcCkpy~=Z|8!%B#sOC6_tT=^;$Xit
zaebrR)8INszJzGkt}w1>Go1hC(uhc+^h{b;kJzyc{DqZ{x}WPN&AT-5hP)tZ^V&3m
z*0Q4M=%*u_M53mIase4I`$kT=O@6dQ9<-v2xV;g}=!ux~-bO541stA2EGNpR*1s3V
z3#bR2`Or!&g#Fv{$n&0_^`6<a1di94sOB*>$g^1F63&A6_C`TQ%T~!1&^CIAi=0V=
z7qcF$^<Aq)_-#iG%}S5%`yIxJK0+O*{4jVZsv%r)m%exDZ0+FTzg6)b5kBR$L20?z
zF(%U&kC+cQ%r(J2Gc~ajgt}unGW?5dOKj^eDOi}(v*zeL-Cp1lji7PP0)te@0Prx|
z(0Olyc08dfOQKv(`Ou$t16KIBg_c?+D*Wk&-XAZfC)?@Bz{0p*TjO%$lJ!HgqN=1s
zF6J5VNF#lZk|WS_*E2c-34}~q?{0sOEbmZkEqe}&TiJ6wE*?fesRoxuOz~;GbH@1V
ztA`|Fx-q}+Ou5*iEPzC3Os^TW#LjuRN{q<ch?hdK@dt|`1o#+9B2+;fps7r@+j@pb
zfrh9=T<7}qrV8-u6Y<Ww$lEB=>=?e5^Ijczv=d<DgG-6@0xfSi%+-5msljkm02Y`&
zmOg8C<g@FgsU%9xhC6?=_LzaY<eX=o$XGpw2V9RStb;!CEWzO%(o;myZ!F8E)8O3^
ztNF7(mEj?Q$?CW8Cg}i!>_vwLzth^%(2xhGYtOz--bn3?$uFEs&3C_lFZjiDnlE@a
zjNAlIK7K4D`&N`&^lm|Ap-mJm*ebE9JxpyYRVM(x*Jh=J3|cBuv^bJZtnz6;B0Wyd
zzpX20a-!J^wyPA*r;~2`b8i22YyEkd5yfrmg8x;-Li^eLRcN6>?7Q858Xq_X`np4O
z5W5Nz;@YD}YxQh265nAU!wl8uT0O?7f#35W_8?38a<fs!AaQ;hwJm192$fzS^dIzU
zWm&gBR$SiKp><60R4(GjBEE?n`^IHN{>{!8u#$LpE)q(a%K*4}{q%S46b~7bh8Z3c
zg;iaSDFoQx;QEfy#_S6vv#aCx8Wqzr)AEcYI`|(m0&m?0jypLnNqC<2Bx;y<O{EYp
zN0!4J@9mWScIyLL-5scjvzz`%lI5MBpDVC22;Y>COb$Wru795@rlSPZ#^I6_#3eDz
zcl&s$&+d$*(%<sz<cs6SN31Zl^6pYAh;O`41CX9>4Mg+WQYYTsK0ZE3q$>xH%oE6V
z%8g_9_bCQ$l6K)*K}ljp7|0=&hT>{iAJ)S?K00;V9A$0Nd@yTZMxF=U-LQu*5Ws6s
zSgxgjLJ99-e}jB+qJ(xRQG>9<JS%7$s-2#)sx(#}5SfW3pvA|J!NgRHhzw64V$T0z
zbxi9^9-oANlcL2Vyj`(P_V5lOHlf$KxF#cbSM%%d(1+9lqMaBaBB|JIj*F0e>8lXF
z8u`#Tzo>6Z$`pa4@i18uem<5sxs;IIb7?q}gaHY)=jMk9*?gtY*w~;Mx<_|%H`M*^
zGrJz43o&zt&wSbY!5dE$KUn|5DEA+DPM;wsD+7-6D5d@5DaDUm$HrZ4e-ju{K21W9
z#0iPIhj1_ra_-__`ZK{IIqcZMJ;pFYHIF?nzbB?*g1amNkHZ}&Ru?JlyTCO*&fJ5^
z1b1MMEEAWg6KIv0_ZJ<p_-{uER$%&H4J^{+5aq;7Es00%=|BNEm!J4kK!8`zMDedm
zlqFY{j8PUm$1B>HZb!e5F(y2*kI#8%K3&mL4CVRklFmQ)Rzmx*|K4Let6chXfdso&
zx)9SCRX*-P2c-f3q_q%eD840xFeDv@Y~<aKJ55vO$x%VcVr+{Hpejl<?N`)JN3qYr
z{K$bOm5Bd0k^-t&I9YdKEEO`mEO-HLL$y-$&fmnrcUtaY?lN0OiuK77<UV%4gj#OG
zz6^?>oWM=a34JM+H+s|kIrffIul8^t&mjlquN1XI6{WYHG>i=0K_{zPh68Rd`82yk
z;E!8aA9rnY`0|Q$`cOhgUue46jpfy8*o6jkPrkK7QZ95eq+(+CyPhvuWxm$&Yt5}R
zw)?V=v40hO-g+W4vKynBk7&##&Dn2n%sT_E)%BSIYXuX2gyth0`dKwJmL!6aj_K&m
zz0)BJ=knIX*Kv3UR8<NMBYk#IYWKHt)CZZqBDVwf$|zUtnjRy)<c-*3H3mqg_SA)0
zOP>IOp&yiA&Qua@FgwFF%3JnOHc*<bJ9oa$r0~_<%^vf565Ea=VCGt-YP)=Tf%pW6
zYoGiIEJVD4rYDy@{`X#pi+N8B{|q8f*DU2_C*Q($!yiZ-1F}pGs$E-z6Cbb0vy2fT
zpVC#*enCzBe0)S{S!KrxRx0-Ll=;c8*l(JNcrg0=(Jp<sN+f}nX!pH;9Dd6FIt4Pp
zY}q0HFppKa%gnOsV3fcFpZ(9blVcOt^9t`Gy4{$EoU}B#>;($b7-bMyj;YC%*M<@S
z{^z3py(KE3XSe1TmwLmBI^i+=uABajq>T+bW`X6C`%azBmx9={S~N_`hYYPlSIaP3
zF*t5OSgz>?F2(8m6$?P;S!K+(l4rgMu|%6u!`esXe4@z<_H4nRx%a$Wa`HnyjpX@q
zm4DW@_LFF`NBm(%?^>N3Ti)UYgy1c^P~bA1f}UE3h7n5)qU*R;1?KO*819?VT|)16
zVT-Z!bdu*u`0W9yul)(n2U(YWVxY2^B4{JOO%>dmFn@oUK@D2Dkjd~aaF2~o8Ny_Q
z_P_r4uA7kJ!yW*>(MGd8>@Ga68xdN(icFcBd{gxhA1A`bgtz%@zN%@o&Al^2x_%>X
z=}e~<{U`eg7m}hi4-lAkHKU-^io#2nTZBY%@P2$JDHXR3VHq4qqWn8CzBC=!GWLF{
zeN-_C9CH62pPHJ?ulV=zF5JA%xb>JkUuh9Xj}M#tmT$JI>x2|DZE%5qOCOqRYc&4I
ze$|yvNL#P$qImV-Ov;9tjSh)87laZau2oYy8yK6A;qSW%jB8;p#D34hYX`gRz<9@C
z_A2U`mIoQFkUj-y1SRqu!NOj+nHMzG4qc<#frTI@PBRC39|XBG1=65?)RBrAn)g?G
zrr~-eaWeiZ#NWv)!avFC7gO7ul2UQUB)ny}!E(~b1FyGfCY-l0E|Ra(gI<*9TibQ(
zFP@fAR6;nxe_jQt9Xn~Ofee1rxDr1+U-E??8!ou^29GhHw|%x6hCr_Uhp;x}C9LEM
z7}2P6T!#LXC=ig{!CNqaHLzjI|B88_?P_nbA`REk{Fna`vBvi$MMCFd%~H)=eG;kc
zSLH2u?BC0#+T0&eo0?7@qCCn}$@gf~gZs@iJv+rrP?#>wE7R$epN>)l{n*X%kv2WV
z_^}AUi&G_%6C;vOhj)i>HaS{4*(g?sF6WPry)i(Lb}hhNAj;%BD4lE3l5NT0jSwSi
zm@-?@6}L?Dd5_FN%<_&`^IgW*o_N3CJnd~>^Il4b%t(D1H8$jC=APR^OW)T(-*#Qd
zx29IwzxeP)!w)u9Qw0SLQCe0FZK;UO&yXs#0!P=cARZ?pfkhvdyR<-Pgo`=5(qEdC
zC9}F@6}6aHQcTSH^o4Ee!>1ZHHKfq!x&1N>&;pYnee23F)HYp_AG*YB8FtzMi?i$L
z;J2?fuMqYRv|~}lbY!;s#x~YrHxo~L)E-yWJQL{WMG)X34I|einXG8)s8-xkxn}Gn
z@_54;GL^`zr}p_~e}7~}3QW=(8*i9K2|=Jlxya_$LNu#0ggtY<RUPYAnhEI+x7LoV
zDdNb$p(W0x*zXKl5~K853|t?CNYCDa*&tbTi>9*uyZEm6vj~dnr@uS=;zGK&ns+I<
z)x@`u2tVqAhm22lwFt%wVTDbpiow}=_7$ntd9dKx>VmmwCUjI+YZdgWErl!Nt{Wr`
z?Z-KE*S5%~5+-S;FlkVmjr2*m2SCYbMLD?}yVqLI)!MSn7Il&K32|oi_(_>X_Romp
zB`*!Oqx7f9-JeScszt7NP=HZg?9S$|jUw$^b`pE77J2jZ_^7E31N&wRj-P~bJ0Z^-
z^P`^E<Olf?`mH%I^<c8yHtq(tGrwMwzb*VNo`m;2a>?a_cAQ{j>v-lc7xUFA>vYhZ
zbHIU=k0g?wNQ^1^o8=(4gxyfwO<sYP>s^>S(r;F&#6UEg>?UDr;jUs)KZkC9(i~Am
zTu_NbKZT-^5cRn+2>f)D#Z=0t2bjX=!}w%Ti!-r_PmxVSBCNx0-y+%{UYY*xcueP>
zjq_10KFUlMXEA~D-*U^W+srxQy04o+TKIer%_GZd&tv-(wYI-hYKHNt1{4?oObx|2
zEF<}ty0@0zD>5$n^3h1JGwHhuKuL=vYBdj@k?<$6A$^Yr2aWS@UC*CUJW?0vpqM4X
zvf*o+6gGUmB$(vXDEEja{;9;*#J`Iy=SE^W_JNv($4r{)!EyMp+QcCbJs_e|TO1SL
zmW63ERBiM$%;m%aTDBB$P;%l2N~j3UV@Z;G#1K2(Z8VeiE&FuS7Ov8uN?mn`4>GF`
zIF|1J$$u;=(sQ$hjmMd#%RAoHX5YGN2d25|e5mLHqF}wi*3oahVF-;aIN`=|kkCjo
zCB-Fyu(2T48ci?6v^yeV<TaF%mVPqduh4?jRda6x2uOMMO-+bRWfx~A$)J$81JMv1
zf1>tZ49Sr-Pb)UZmA007x5TF8H(vH6EFcP>oO%*sx*r-p#Z>GTw2qHCC5dCY;=O*1
zf=ey2UH)_sA3{TlT$2h)5EmsGLlQM8=C4nb3lN1(e~%NLXCV~H$0j#TBdbj9p>ji>
zvs`7>U5KWM<THbw+d;J^)cF;)@=k>%45iE&M283C5m9Q9W+U=#4BzPXG-G?%LXX3W
z=#has1Yv(ZBaV@S13_|rwf+o57D2kG&z~=0BKkvW-ygqlYZ2txCB`Ta6kl0f?w8vv
z8-$@nGyWz5FDFb)3zu49o9f{S;|QO#HlVdR=9WFG4u2I(UEIRgjqVgBSWoNx>=keD
zAoh_p32vn42+@(~Mjn2DgWe39CV|L5^(Q%Mco;pKWSu>2xIBi3>F7+fG}K$3|E<A7
z_TNhH;fcOp0`p<`Zn!lj^P3s;Su0%szua>Ah%4{S52ax0(^v-;%mOa!*5{77l$Z1s
zybX^@EimnUA0ZVpNiEJXB;H%I;BLlJkW5eRIHVs}H3j1>^msJw8tV8Kwcu5yhu}+a
z>?n#GkDh_{dEX?wD(U4!cl39W$}dF^ny<+WE=?06NO=4c+~M=`Xhay_+?iYDipy?)
zMGN=f-7tOJ1h0Xl<QbY^mFP>mrV=!f15>CR5d@d&qt@zI=vXYYYct@G7w}b(X|y>H
zwwz}InuFsWq9fd;=ZWi;;kX*=ufm=2;%U!kc;veIo#8Zs{X-6mjR)D_6_fXW+Mapl
zj86e2TwR<cF{=Ah4>RLegy2MfPok}!ijUAI#J0VlW)G=#QmC@A{mlo>!6nzHx{+Uc
zJEr$G(lsaa?83N#KiabH$Jy5tC<uKrTLyS!KPp^4E8b!eH%`NTVLdJJu|SIes1rlJ
zC~Z0JJi;(FVLsS`@7n2*5G`8?eLv|l7MXJ$JJ8n<mAA4wV@{Qic>QUUo;^a$tRdZw
z-k0our!5OVJX&nf&yZr49*$i{c(d%M^<k97k?~x?HEG&2ZSKe*x#XFiEQ{8Y^)S>Y
z=w7bxH8d=HUy@Xy=Du^it3Ay$z(I4>6j?)5qx_KR-*ne3Dj|HE)NWR#RP900s#>K$
zo#-NnaxP<;JuN_NCwS_)xq~R8JhZfX8ib#_Quix?B$ers_vr+MC%+=BDk^frVd96`
z0|n_9Li5ogI|+*HC-LT?%a_!?41Rp|b+}CD)IQzrn|B8Ud$$#SZ?_rhAMURS9@wGD
z!K2+bay{@9IG^Bi(>&d!fwTuLyzQlGPh9xvu-@o`FGik2e7jCVs#o_16qv5+WeZe~
zPadCs*lJi|6p&}PSm=&rF`ze%ipi^XMbFpzq{fX+%a-sZX_^A-zzaGbsk>OG$c;zO
z&&2iJ;<5#AZMmsn3j>Gt%l5#_PLwRD3t!J>L55u6_u#nrSha5XvVz|a8Ovw0bNSk`
zjjI}zFrcBPph0!X%K3)q5Yvq__@~u!)n+-Xf?Ra+*>rtF!9ix-{WNkx9OyO<DBza^
zr1#I1y@^Xc;{f3!_v|pAumvLm&QTCaA~X1&z-nOk#YZNf!%Iz&a5am#c=?$oWHRER
zskWHQU_mYdfQ|pu5Sij^aaz*NOp1)kq{OP>-L2q54JptVWWE?ZWEz)uq2$`B>c(Gs
zPIl!2vGq-fqdD>^wBeU{9eG<)@Ji@zEn0YV&``F4*g~es(VvnFW-=ahD4kfb+qlqn
z;?LF**#2&B-<m3A+UQcWhniCx`gp-^1|IkbBx@C<9;>cRGrgzqv<HCrr9;N~**dY>
zj5*4EviL-^wT#_ub%5tza&Qg+(`Sbb8r`dvS-uXA8tc!>Af|}kpOF9Co4AFR03|w$
z`ihGE`E+Mn1EE0g0Y_SI23Dfw7Z>K}rS<JXKHIp7RHuR1H<U^Y>xzsFs}z1?SWGeq
z{t}$_kTYl_yXlzcET$9lxxSWmr#v!e`>kDjfg`OG`kub+b9URizNOz9{J$AS18GM;
zOygral8yW;#K~o(&9%#RT^8f?Se7A?N*->gy!fADL5wYa&@Fne8Wn*oGKcumZ4r%n
z$?d9qm_Dt)Y9dArIt`#jo9PG|{6+lACj@i(@*ypjjU^f}s+guECVd5g2Kp8xFrJ_4
z;aaP!Bv8;{i=c<&qR%2a-X~@;N`>I_>NoauB-+_S>=-Upto}y7+IiS+%l^rcfanj5
zp3RBhk-<4qTA^dAjJKUSfMQoHs>TUk!N7$rcVtqbyCN(V_!xK3j7dF`^b+k7QbFWz
z-cRd+duO$bqrrKlH+u~IQbn6hn`Aq|B{C{+88M0*A95)QZNm#}0-?_z)rJ@-*go%h
zXf;5^N^Z-zdnL{mA7|!zLk!-dt{c+;YB1#-03Vejj~IIn^|nvUlj*-|RZd#WL*=#@
zv0ZUFV0;wGDe#MP8rgzwGdj}6;<``Ul>WBnqoc`dQGYxh%XqXb@w5eq6UbDfP(ETv
zyy_vPefl;CKV_v$ajQy6yovdD9iKiUDO`b3uE07tZaeC@gTJ|+LL*Wh3Q>_TEYNm+
zi)0yk;It<}(si`QL43|-DbYM*=zo*FW~v{xZ^mMmJ|7t}>|Ds&;qf(529!Q46yk5<
zGtD+coA#MslQ!}&R9W-6+u9vR%Y%5vANso<AQs^T#2N&t9nh~7h#!@CEd|E7^cJx~
z&Y_{d6CLTcd=EHfBk*wi$hxRLgp@Hl!-n}!87{~woP<~(X>_7j45RfhzKzn>(fpcS
z$Km!ok{>-$bC<he6+ekn%W15RhJZ*)3M0EvupZ&-Ux!XmrF@-F-O7U^cW9zf9o~vO
zVjxQ>cJ0O%v9p5i6Lr1Oo%vUGqw=0!cF}S0`C~iW2W26#_Es>NXUwRiZ4sKxzW7qH
zQdxGisxc_}r0Iya?>nnqRNTKv$Eb!JiMB`WZ-@%;EsGj-_QiVo<Tn*F$dMr2*pXsW
zd(2M*hsOQ1wZ<<P)Mh8e7(5=Mcp;ed4GkdzDq0Jcoi+?ibCD(_xZSyrRzL)nA#UrI
z2Mslm&h~6m_hRVw<|-mULF-fT^H?Tlah*Mn@htF<?46mYUEgW8gi$ZJ361A05oT0}
z3NpU5r8O8Ap`0Rm?bP^iVOu`Da8X5fGewO@N%4r%_l+w=WJP|g0~7~8`CWFd(fwmu
zx}?@HG!nPEV8KIqYA^2TBvqAQ!I+D}Hy0L;Cp{EP<z~L96OZjj_BPl3nqV?gsBmTr
zrW5(tVEDTmV{lA{OwpLq$YG)gQ?T4dFAH-8EfLPO!au4I7NjSyf2@T;<l=JWffXj7
z#Q7!wEKrFeVW`!8iQh4#X6C}1T$vsB4iEI8vzy}xlFUuIjjlbj&ICcn>S8{*_n5a&
zpG<7FQ~Wbn3TkI$8e)hq$3J_fp)p;>yYtgOjb%Erod5TtV6k01LIDH*PuV;}S02;k
zhqDfneUi?kM<n#w?ZD+rocvy^uQP%B>upPo@^;il)#%r3**A-g&ezxOLk~B){G~P`
zhwJ?ON1z5qSR$6soiyeUQ1p!DJ^L8<Xfc`3S=<BPnZEqQ8xBkUI}XIihK{y2X)}uy
zD?Hvgu$4ZB0=9a+&3BZ=uktrav?=Qycc9RsluwUJDV|~6N2b80Lt*fcg);-TC;E2k
z$1qqiiCmQij=Y~~k~K<&^u|p1trMCUnx*EUhwEJ#0%b#5g5&+(yM2#BCo(2~F^RV9
zmDQ_aN>{MXvsgR3>znRMGb-Yb+TJT6?$3ONKQvh)&r(GWk0wIY!`bovH1OIsa{PJ#
z#Y{v&!%bviqT~8tTG>i^i%BvXx1;mq=P<)NOxghEQ9|BG5j2XEL`;dlKQb`8t(LV4
zP_Q8dsB`u$yIqPQGp5dPq}6yh*EzvYuDpxe+7H&TYWB-5b9#rq<h1Ww)U}^K)?oho
z!aVG_(~8NE74nwaMz4l>q(T3c-5n{I8@V;2a#gTzq-cw}*}fTXnncTXwZdvC;Lbv~
z=o3gZx$-l7itc$G_a!+P`p%RY_rcuagZ@=I^86sYYTi@Due11K=PGsvJN+I<bEmbN
zEB8RPZqcvGe~=45rE*X$UyqqgU%fM1&Y8N2VVV9bUu6m#{Z42>OX@TdkHP$6aAp#d
zxa+J)x_CG?xc05~V>c;77%r~W+8%w_CYt=y`WN7%JB}&<XauML=GoU7id0x+taVX}
zZC&k}AYqlsdDoXt>x<eF<@EOY5=W3;4SSxFH*5P_iZd8i7>-$*Leji!$dI<VD&~Mm
zdA3CwyS`KQNA9s)QEOC=eB%nu*6OYu^<hkjGSR9SA5-O?`}{AfrkTK%bx7e%vDu1f
zlbhFlsj8N$r;GGYGJ=Ib;j@p!^+8x~`5S|K<b(^NymCoWE1+AL{X_}lFSd)VJ4ecT
zmb{hcSI0T=M*F&-GPx9Tv!}vm5Ki`Zv+Xa9_+d1+FaWm0DlEfPynaWLX!e^)F6hg<
z8jTnCPm6mFbiII6)0ZGb*~4HQ<2ZXa6dmwR1<2$#YUgeY%d`rTE8R-<mG4-igfAs3
z>iq^oo7DY}l0WF0K6hYFk2HgbF>uq9p6m=uRp{0(vVWXgmx!l@h>-lI!C;DhjYs$g
zv&33|j1*3Gy5#4}Rv(KidCs+$Zsak;T_@Y7d^CvZ=l}WXMBu{PZQ6@C5u;z=*IAXc
z`Za@*-9q<9(~F#f$cC@L*2t3+MS|$K=q+|Pj7&Q8=fU4nZ+XuV|MJuEJJ|ROZY=!N
z;jYLe+c>UBmX9m{7fWa1)`a)AacPi{X2|FU=^EV~QX)vh=x#}AMvd-LQlz`21x7cF
z5(K0}y5ZgTcfEhXcJ0}D>YV#?H~zvB>@kbBMIh<Iy-PIUX!JfhKOi7FZiE)}M<uI*
zT$O6Zm!=p2)fVS<{>69yUcP635)pH@78$mUvgp;C-goP_w8Z@ED7BnH@QvS6r(XR?
zRK$NtrYtsvu`gChCFYigZ95a+`{&Sct?@0VU2~bI<0HDL1yn{QVUPE8E7#SzZ~_Q<
zq`}*?e*ZDWewl0$#ZcywFuJ}k+&^0yXc2`M?4KM$%_?!-G);#dQ(T}-)aFoOfX|{0
zCi#cvW|1)2kjHi-W|7U$KFU~2Nl}<$iLb;AbNO22Swns0^K(6q8wH~KNJ+`RD8{x^
z!jI@J&#0iiYYI`ZfF2({r3^VXiFB9?KjlYt`@h`?I3tAmm#75Nip(XviQWm9;tPHK
zTDJ9w6hIBWm}_W1v6|IugyKks6p_g^+b<QiINkM$a0Ow?u-6qp>TO$Ks3NM>xQEK!
zJYCS1a(0XBtc=a4dVIQxXyI>jE(Y8F0Ntu@9VhYv3)adO3akEw#*6TYpzSkl4`zsv
z+flG$QT=D%>EbN{;H(!|<u_XuyN{Tznxh)X>t#z~fVZ}}{?j%jmdoUy7si@E%M()2
zV6*PLg6`8SWZ%^wwuRRTW9QOUK&m`aZ+x%MAnXa&yw^=aavlUUM_eVm9NIQ$$xg@*
zsw-_0|H+*qz>OZMX4WJGlGfk>N{YPfMq?2{3HlTGtKN}~D38AXeO-`E_3Vxh4GlT4
z_>QvcdwL*k*OA<-hi($4VvcArx8L|+$E|i;N)diF+Oy?|5K;k$TZ#j3h<<IQR->$&
z^gp>8Nn|`x3%U^9N^0ZZDC<g|By~z}^NHCBOhhySScf!-<0s=Mf|%)c7Ss*kJ6(~8
zXHTdWjcO54;-pyo9V2d7T&Xgv`nPv-pQhU1B2Wwn+q-H?qVT;7TN^XjOY<6u<7LAM
zm<}P{W4d9p7*|-xM1xA2Szwy+jZrbpI~BRhnbi4G*!^u-YX_+!vKGiOP`IGiX33Lf
zz1N3Fla{FO)Eqy!b48g1{AzC!<Q){|Pu*H@{8~^7(;a$ioXd9|KM_~mWXs-Ywk5}R
z-k89OkdM=H6z_S9ftVyv%A_z3Pi)K?RGFDeP#Ba@{%~{S05K=@mfwJ`Jtnf)n+*^;
z^i;wCQW2#>ZfE;y;?-~fYeOr3WZ=Wdp@KlD*_J{{%JyUHZVDbTvkL|q2=#y&!o7jH
zN3;5tYJ$8FBgvwrjoTq%FKUjI84kHzrgOj^3iBu?8k)E@X)z8Bgdkg<dVW9IK)r>w
zB1x^*of+P$Yg0gDPd#G)3V#$_9unR*a@Vlyzia5GVBH2pn;=pn-<5pRBUu=8p{*J~
zs<?jqHPJ*?Jk2zLyGCnQal${XM;FPVh>Kr=2nVcc(3glQH+lSW#%hl19)E@q)Z(II
z=zQopLgnlx%fd5Nq&qNCSjva?NG=h3YPp%N9zoF5W1}GXQG}|H#yNZ_w3|7nr|m1>
zt`UBcCkB$-NV&2*J;ywP+R1@n@h7Af71U3n?fkB9#;i6AvjxZQks}uzDx>C|()uZg
z7n3Na+pbYOr}d&OhQK~vCTSxWijd1Qs{hJCSN5sv;q<FNk*$RVe*xrjn?9`>x40)=
zCoXZX1yz)edipsFehD`or+<oRM&NRBWV0hIz9mY&^+U@g(ul=|tz!~4&$13xt@RJk
z_i6voX>{mWnx;-7)fE#8)m?o*$M&{8UR$L$$-?YBc#bqeBY(>SEiz)dPbi%eS5Y}G
zbsr?m3_(l(+|w$wcKKB**UL!ghhfXm7|BOL$(~i_uA{@kSQp;@$O5WIs3?k-U&)eP
z8Bv(SU4`c5?2_wmK%Iw)t9J45mScnlGi$OkKZA25IG2Ua!XkDk<i;bg64JjvSfXwW
z9}=;(=LJ-@HqnD*P(uTc34I)asxvdRa1uK~gzoQu@fAkEcod!=$2-Rb;C5x?e+2$4
z<4^>BnqqCNkseO5zva^I7o=2QhF7!iAcBp7R{%red*`Iix&i?(q}g#Np-lG)Z=c#;
zheTsS*e19_BOP`Z9ud`{cDjef^V&054z`It70f+UwsFy_`0vlyeenf7+tAs@7ySHN
zBdDqf)uV&a8dqMOXsa}it7GxWv3M-UxwsrHW5=buz{3!zZm{heD#U+1Udnx<ibF=X
zmNp#IHGF!dr<x_poCNvAKy;%XKcp|?pU}D@^=7jn{^Yy&17Eq?k0Kf1@TfrFgsSq`
zO+07Bh0uW0E+76G!g??$b5MTjOtfvtsE!Pg1maI<LOh0gfYUq66mu$#BnYQ95tX;c
zw0(QF2!@4klA_Py+`H}iExmFKIwWU(&})TJh?WX7+{VX3Y(bjlStXi>rRjaSGPahI
zgJweLw;3LMuc8Z^R5{EXzwfrM!?eW9ciTL;Fmaw5rHY?AKU7vVFIQMEu0d;Jwl2&~
zV^0I75SN&NRo!J-Uotsb#6*t1;<I_aU9Byd@q$qLq_Xwe3}U^2w8GcUC2~<Gl6w{S
z*JO9em{CInuiaW`7hG4Hzml2{__RIfZ$jhO2^W(LTTQp@*WTqCPVnqzpVIlY{!K65
zc$q#~zC7@9p#eSKlCM9<u0vSrpt%o|%c<;Y|A=zA<j8_pMciKZGIIb%X2hFqm%Da6
z_P75LVUivMwC<kS%rvE=CP(Z!G<s|{#mCG<)4>Xx?Y{BFzs5vQz+@|Q&c(K#(0(&?
zI#k2s8vz3j4XC{G%$C#soUrZjpD=3*TC@_KO3y;BxunDq8tidY`Y}+!tzuqM#UdW@
zz-{9#4S(Sx<avffTCX;V*5Xz0_`g?P=;TJ!*NWRpPw(`n6<+h$RY^dJDCow=eU`Tn
z)Q%Nf%m40SPi$KM83&XxEW88q86cg|9tcMxdziYqX!~e*GqAxyLa6$ylYd(O71$C?
zR5>U%;X6eTI}$H#(QJx{(<I~P<qgAvm$4MwXZ+o)K{<QgWD*%okir!=APB3Ae%ZQL
zc=PiyFU?#~>cKCX`0uiR6^o)lGP%U-(OdK~m;k<Srg41x3dlk-5X_y&z}<42vD!cW
zS>I%tF^aZO!g6l^{U9%S^RVpPQM8POjOd3<Usq4gM|<zc;Nohi#~ggx!znCB%K&Mi
zHlE#`u)xuFdjn~Wb(*J1y=H=qlSgRW!-65rhRKM?>i#Gy(%{byg@kzo4Wxa{YC;(G
z{1mhtY~&J|#70Eaw)Ki2*XLKLlSc&Mz{)#TGp9Gbgn4c8)SKB$4kKNwv(vq#wL#<!
zzsvb7d)_+vC=)u$ov$qo%FXUQuI*u>$k@qXFG@HQ;&sJ%p+rB*uN<Ax>>!l-g5eb1
zQ7mtE(I2)%%d@sJAz$HMM}<=wBU!!Jn^0Too58+_*f7gX>qs2oS;xldGwFw`wl^{r
z(~=5^RV31%KHtqp9+nm@Z?eOT)^s>&3TiF$J<Bg_Zw1NCg~~d~V>fnM8oWE4a@;)~
zgj`?XT3%9R8Ns-Ew$90sx9~#AAy3&5-756E>~!uPwLxh8o#2385^Jgpa&XbV7C~f=
zJH=(;vj6_=qdWQ?<;}M22qpf{Q%-yK-eLEQ4yh!=Y3Z`;xwBeop}INr1y+Svy|$jY
zzpV&6VgUDN(nd9%$4@G?wD_*!-@l|vOkkKR*uUzR1GL+5VdHT+2hu4Jd2&AL0pAL-
zPb77d8}ppg1gO0RIpG>p-PYEWmKMotO_hxDS?}iEymo2j|K}(s!c{)rhe^VW*wa$}
zXw1980fM-5-MN|!AtDIby;jt*V-ZLDS_Gq#k{h>0>hcy87bFpsGiPniCRTAiYoKzh
z1<e!W`-wi&^e0<+v0gjizQ%U0Km(f885qy~M}kgW@t(pIyrRzu3v~K+@vGp^8{dSE
z%UA#MAR_2?o-6^(i!pebKPW;OdKHE&6Hp|jFm_6HI0BfHcptKFcmPmU%QJ&w-~FFw
zqScboqErD%fg?x~iI~)RqSG|i4axQcJNt6~Jx_xtB|`^o&hB#G9K|gcSJbpEU7iFg
zic(m%74l6wn#O)$?Rpsct}mPW@8KBd{fPVbC-?RKwyo8vnsbwT!ooAT@A&p7T{wWY
zX1z+Jb>(f|WehkHmc?MHVo|N5f@d1a@2DPzO|-{xK_BXSb)oyeeLiP-0X~<3z#{%7
zw2uBSZ7a9Oj?9GW;n^OtS!s<u;!e%m(H`*$TdNoC*@aaIm)z;#-c+`1<{xs>?0=M<
zb%EIxg9sr$v*;&oa<pBD+lm%rQBv_`CF;J8^#K7nJwpz0ni~RdF9ouqqkF`gz$m9z
zAgUzkKaG292Lx~ldGHIU)p523w&2C`#~$Of65JjOiigg|v4>9}5PnUE)oR7JKB6|b
zo&E@|OWyPlmoj|!Kg^$}w>E}N^fD8@#LPb<7a%g-s#69{L?nG<&D}~~I~7I&<?FDX
zC8e;0;)lp}2CNvC)ed;>+)m$^j7YjkXZ5kGvA!OnBy1L?rftlAaeH2keueXS^{<ZW
zLGb54Ui?o=%nafmf+xQcqkMB}ag%hXyu<OSjjZ`b-&g(JcuQQ4Y!&8!@Q$gbmPp3@
z(3sN9&GNUwNi?i~4`tgJA4zWMmg1#9aERwzWdC!rBh<rY5wHcE;xbzVD<fzpwK!Bi
zyUlC{aVYYC{<hEySk6Gug^zQE(2tzgLjkhznfElGL6f=ur~&*c3LWfl01)3RC7qs*
zko1ZiUSaHvzRrRWu@h=h^m*$`g!Q1nOZZm<KPSXvwzt}8A}*X=SA-REti5OtR47Xx
zC~l9nw3PVc%`kVf;*c!x#CwG09pF!(-CX}_q{YHd|BGrG1*No#E3L2r1ka8*Ywm2L
zRmnibi8F4Z)fFHN9j0_EMoO}Y53_i$9q&zy3ShxQ4>K2$?8DiT2hrjwamiteN1rGW
zQ6!uF8a*xPLPJSIWf-!H<ymX2JJvj6`F(OqM16fVQXz>2ylx8#dwyS0Wa2e1gz$p(
zj}nNV`$-ey2Go`opnu|*iSaO2scdZ?_WhgPQT_^XxXdG%l%O213!-P6Yn1&7#NCcr
z_q&jS42@6+A`Gggm0z#=R9+2;PqdZN0BL#b=OlWYJ~v#oyT&7uBXjG^a-op#=IEC*
zLzW5embTrTQ~?6o#je7AMG@ACc>aeJ|AJ{ruX4FG2#!)$@TsS$61vYVbpRlb^2R4w
zs&A4OZw)PHi^ePZr4ao`>KDi|X<@LXN_Nw0Fv+HMKXMpJM?g5L_YjT6o~CbGVbGIE
z_3j5O-z74O>z$layr7rAA^1V<A1Vpvzj4MMZk3i}nYeN&MKw^{NzH68kixBRnh<;O
zQfYFa(!R4RS<1k5VGuLhGRWF~=E#A<U`zjBpdjPOx?|BeuEIZ)MY47Tn3xCt5NsKH
zF^dB!7vFuZQ{6>aR2`kl!00ysnjQg_vkI5vo!zbIM%dL@i;+*D9hbmEQ3O-08neb+
zyOxPPg22%TQ|7OB4)&dmw&Lpdlkd|dx<g^+?W3=MNdvNo-<K~_V$F#)F&T;BpwQFG
z2!dgW+<uC?QG3M;!AiN8<XTFsa=3!H0g{D@1Z@w+=sj-j1Dz+oaD<S9_hLoY5x;g<
z{!{O$^)_rW6A1;%-LI&ITRrgY8|1H$w%UlxmVgiyB^JR4XDR1T;r^beEWODn+VRlX
zCiQFCT=~Q4y`B^%QvF&E5!Xcf>_VFK!Xhnn#^T;%Nc&u1Kac9@o)W&e<y0?bT6b1s
zFr)}S(_>tCchzHJ2rAC-lgTJYt4M+fl0AbhpAq}!fzJJx__nhwm`3A{!ZeI~*RSJ&
z23m+*Fu^X;3qc}z(CU&xMMc_aHr%US#p8|P7-1IROpBn|`A0Cu$stfTD8oEK!@d73
zg+FbM*mW6#xRkvWqq2y_2VMR+D0dLJeM7+`dl5f};X5Yy`xeFH5uvTr{$VEL?7Kdx
z39@WnyH-O`nI<uYT30<!pXa|O+&b>%z=Ad=@7CnJKs6B5AU+wxPxcV$bWyDV3pB$<
zYYyjEFt0|w`Dj7qK$fQz1Z_YG0Q)CoK!r|J?a~lB-c=eO=%@feN~9c9vA|u0kqwbD
zcL^(AyA%#x9wX>6uxv094%Fah)<yrTh_7*!YGz-ID5|GVwlwgnm!5i?{JB~qCAlpJ
z$MC9Sf4<Y!Ljb)oPZbp-wOZTSkO=VRxRoCCRlLO1O3Ikj?HlmdHaj#Yr=gs9uW!5S
zJnd{j{gH3pi25lE{<J2g6rg@*C$%eYW2MM$$3*!`L5@MgKIN;W9Gurb4cS=X^aK;V
zEdH~n;sLOR<W;p<KWxk1;`~>J*_O{nYul1i!Bp?Lp_?|ux-jO1F8b)bImpWa6C~21
zU2f~sNQSXKJPt)kJ`%UXZDDzg{H_T>VP^1}ffGB$!Bp3TX1<HOUY8CB7+NL<&w8OT
z?*9MZK;jnHh)5#G0qit^A#sGA6h7tNLhHOAx95s*en4SO4#$s$_C!m#2jNQ>GYTgJ
zkX|Ft^|U47U=!4Ucy{}mXg@^cEv0f$iiaj@#4UPCocM2U=$FeTFTYcMZ@lqS^NaFv
z)|^Ih^hEl4{1?V!6=Tj&!aAjbU~!%h6!W2t{usr=EIA_<W`sCIaJJ*-`1sdywnG<w
zV-rV73WhIy@%GFHZD=uQ$Fs$G5^eD?bYau?goG)_pn$OigX7j|m=m8JX9BUfhe-bF
z=%|TEWe<}S5(f;#2;PfVi+XdR=@*9LIu;E0w+7)HAv`(|ew8;v>Yu+A{1rl!ZZ(xa
zL<eFJ%5R4P(7bqPvmwa5OoB5PL>!N2dl|S^nb`Uq>${nscFuAHnZ{&e3%-^kCzqT3
zLaKlPVgL2!o6|?lUs^x5kh*l1@jw_rB#mFs>QPQB`>mN+ky|~rjv^`01Z)*d0Xs@Y
zGh%5sZc2m&6nb%8VZ$)PZGiURNnhBb`f@@y4+yn<Wt#hw>tr4~+;y+n<QFr-ZF8D3
zS9~!R!b(XvUHIa1s?p=DTGQUMn3y%9oAyJugiWBBO_}jSmMrX*vLiWR1UCzV6f&3&
zB})mQhKx)r6-wxeEEmBLhm<Dg1Oylo)E1C-3@B#ny(J+D4ERLdL9${E3JfYculeM&
z=5yLGc{ky6>f_S&AojAG-*6-0-+k)0*s<`X>Z<FZfpo?<#~2m2u=ElAF8xZL2^9;1
zw(D4^hl0Ud9%P!}1)2Kk;h^hXwV}ypt&^C&`_8;kskyJSnLyNv@t4^FT3d+30xz2D
zeQv!s-*a4<1K{uz8@*xoq<=r`HvoWbY2g^B#X&^%Hfn)ua0PI=71vrRl7p{U^5scn
zsiEsyRnw-Qu#JY^i_&Q0cj)ksC{GV%0h(ho|0MF}(a@uI(&~vP)>>K2GDox(-gb2G
z)y^fk@3tmd+H%hc-4m=Mj#_32oxbL%WXxw$Eo%Gz=!<t;5Uq=f`SvX%*US05x==UR
zs~`99#l>&wQ_$&}$fWmx)|IN}=ff%&JG>W8u^pa*HrgMMxBjHJ)x;yz^IJ%g^tVSt
z9JA7%4^MkZzJ>{C$UY>T;WnNUG%rySm)}w*(I+$gpzF6){;i99*;0l(%iHAla|v6o
zwigoEb3_&sWWj4zObrT?l7EN>DOTh5n}fe#O|Y92t<kd!G!SGe#l7maI?mYZ%pg(F
zmx0TbFa4c1S5L8x_;$t24In&x5B>oViky2(xPNbAfok>RWKKbe-+cGCF6wf~J6L4g
zL<nKZbF4^mZ-v>VoE=d4%N@I-!D1MK(3443AHdz`Mj(j3o*7>2_WAO4o5ph9MAjvp
zzJ_*r2@#)&ZD?|x|KIoA@hVV~3D(Kh-Ms*`bG{ot91Zo2zE{DM^Gn{4$JBMjUA9!k
zrUa9OiLfA7h!QhpN+9u){VD8@=bsW*R<ctytw+5O`XF8SIDot&s6h5l;yoiQHSHh^
zCGb*lehK0iAO}a4VU=7akJ>Y6vErz4JHGiYbqOyVyC`w5IR4rKX~sd)=AtNi8nm&G
z^4;*goVbE~Hr64rrI67}$dWxt*qf$2HJ2R&^*x%qTZt^TnUFGgc)sR!Gie@?@F=a*
z@hpHIl-a=T*c~kG#@$fSqo!(BTw46wMW5D<#ZT{QR$&I@y2v3_7+q=#LhGV&4w9vt
zvcGwao^s!@8_p-7n6mg<Xe3ND<iOH-g?U+)c-!$3-2r(h-B&T3H}3CmsEo@vmW!Q=
zF6t{o{|4nA4@y=p9qx`r2fgc29m@##n4b2Ppi<(x?#pLBy}-w~ecMW(0bfQ()FmHo
zWID!&+1c-LUyNHNBN)Tgt;-c&3e|5EFkQNz+$hjCeu5ZTwU2craz{|n(Q9j@m(Y`-
z&D;x5Revdgmn`%{-1Ze+33TVdZURZ|1ZaU16C8Wt<J-K`7f&EpV7Ph#NU%T_4kn5x
zsvm$YLh`nLKHncs(lt_cU!6Wb4eqnJt-B3T=A6qgH`dqjrYP@oO{;I&yi?qblf=+P
zK`WrliP-?j`w5GD{N{wo7R5wl7=~9sAV6-iL@igCfutz44JN60cQFC_{|Nh@8xj6H
z3X?s^IMOw#!QUh5APB4g5D3;rrmg)!mGMh0#?LChF-7}<r&J{6oVrgzdn}b|)l8V0
z$&4vn2)i3c8v^Pwid|ML6F->^p+(jk=A8e{=0@u1ONSNHR({|dX&(uk$E%YLN;xIj
zp|DWisRw!;=WXcZ)F;Jg<QV4L<fc0^ER!H2yx4VO_%Xxynrp)KcG2jf*K7TET@pKC
zOk+~Y9ep{^>xqznhx39;104Y+te=*RmmyrGi_c{g;+j0acQ_?gBont&6%W07zcakO
zjbE=94uJZ%`@FB8Yc6)W68l&21OVgad_0{R@`FC9_i!W2FcT?u$04n_Jgpl3KA#xZ
zpRen9JYUyO*!PdRSt8kuuibYHR3vZ}3+D_JjzKsW<eve|Cs^6a-g+7kQoIJo<CBCd
zUk_^}R)*}EiOI7M6UNbqwj8o^%k%)g94FzDvzpNGB}JEV03?gnRoUtHM8h1D>C1&V
zwNDr)HQ+ESL%ZYeAiO>W&fy@Q1y+h_A_nHe*L>AL>g0*<!`|cZLZC@=ecArciA9K3
z`LjXAeReiuI4&T3+RbUyftC%J;||xkIr#2G^mR)9BHWUmniu@bs6~!RiIRMV4>evx
z^uz?LaqVl(=$PLimvPhk?J{=usQ7J&kR-sBLgYaB`>*hKVEHlmtcQCIKJKCM4|0|i
zz5+}>zkPQLuN=No*7Mer7F`UYxwZe*N?49FH=Qgl@O(YpOqKbY{tmg<@w-eezDRT0
zyPeCD5wyerBqdT$*#aTQq%lZb3Sj`$5P`xtpg=xqt1g;sQ_LS)R6`Sf*u=lL`q}*t
zd%dasPoz%`)#cZl8t2X{_Qj`w3+s1{olLxrO!)k43G@c{YA&11X^DsK`R%>046^7P
z(?j+7^b>rYPyW6<vWf@gyll7Nn-Ae%C7!U(2+G4{76KGt4W)~n;{6STwf3@3DT4RF
zf2Ic4GX#A*e4K>TNTJc~OA&#*%&?r?I|V<D?zo8Ew-lC_$r(TG#H~%6$U6{kT^Ft2
zW&BHL1aXCrqgGTdiZuQF<pti8vP;&6PIzTzhQqYnf8wx5jo7(<2T}~0yf76viSin|
z?dah9@@f=$)K02e`wzy^Dt<2|y<4ePH%~1AJmg2`A~X7!qU)0{d!+EleXaNtrU4KB
z_Pe?EnFSI5LEmDqt+rL1d?ss?zA0naD12V>eIJ&4AKg6g@L$fsBPyCwv{1(b^|y)k
z(UZ@mG;G)H;eI&EdtXFK`h*wfEFnF)U3#`TbynX~JK~xo-O}?+Lb?t-#Py{0z<OBW
zIElVWg9gk>*Gp2mSUF8vs=N2qpj27;VtYXnB&%+j=fkUz=PdtR9DfGv5bi`-AO@iS
z-B$&pGLr_)KUqMSkuU{ldPw}lt)Zq56TAUena68R5tsVy^lSRdlJIf5a-Bw;5$vnA
zfBK&PU}xys*_bXicxcQ#afz5XJ{`B#X5l9-PuFf5ebZ&`Zk?~AU64N}a3pA&zDw0k
zJ3V@x_Y|Len`jAmN7tL=M#K+$hT|u%<pv7v0nLlT1P&S4e^>AZv@%4sEfc*HmldLO
zqVLli{xo&^*C2m5|MDnHaKRpnaKXt!6N{o&p4JJ3Wvmb^Zu)9~dhs;{g%eI(1kbdy
zCAcy!tB!9d-YUpLH`k-DG_LQJ*(gNav`i2UTeIxh{uN!Hrg=+v8LBLtNq9H-D85iP
zsr-<f=a-FgG=|!vmL2z@-~E{rAPvMC=g;%G)A;^B?=B5%Iz^h(u3;&;J{}6hjhi1a
zH=X15FwLB0)Gs%D!D&B<kA0DVxiwj)OdJP}p<tzyuspV;neoR!11Qq;@F7=R5loFW
zsoKDXURLXiWT>OVcqVN?JxS;Gr$4z42c<x(LMM%AJRV)%37oIZ4lN#@{XYr+?kwH^
zn7pDwB2W5|FcE?SK0ePoY6vMM6#Q=Ktpt?mM$M=%yUQQzaGQTKQE&3bF@DjGd*`af
z>^H5UM1u}QW%~`LBJ(MUFVp$Z&TtI0I;D>8Be;S}v33u0mySXjqj3oIDRwAnIDOcf
zyt}*(LxUT<(*BJR>>&r2>V`|o$>RS>W5OIt&0BbLK5s%U*Pi26o+ElLR%4MlqH-E5
zyTk>?qLoh^rl6JLl(F3#<14ruQ;+s-L<btlX24%<T^BQp@1KzMibyKL_p~X=oW6Gj
zzZ5BM=UIl(5Rjh?iP{~8JIO6b={Xg@gc1jJ-CH<S-j~18F_=SS&7>}xwjCl<?xx<!
zCa??6s$`&tq-Degxu_-RV73dS*H@||c`GYfzoZq-?aBZnFB_!2TQ%JjK@Oo2V4_Mk
z*J2HE*D%wpwEPDo4=pjkfy@lQiknC<qq{X`PMj&}Y^+12?;~}E-iQ6<Y27Xv8w^od
zfd=t~!YqjcJ?4PSWmA*!m^qkX1mB17qGgE$4fF;@*Y{O$Nax&Xg-7e5ypTa7g$22z
zaLyZiPi`9edgRMo0{l82L9nD^tHa8p!iLsl!bbMnqy>Ln$geb{OY*Za3Z2E!nT)pd
z8{n10aCp(qzu_PGD7+;tiI}aw`!jVh%*RT9mwOTe5!gYUz~k38-`48PVZP%g1}g)c
znO{5@bZey|F*Ud_*k6+7b$5PWhzYGPDU!GQiFejItO`rZDv<-eOrS5N$Cvjyjgw|S
zIHm=s@nw%7yliO<9OrctjkSh4`vWUkONDax_ueM@TBa27Ic9s)eCxs{ov9H^2td`6
zpne@yZ#wu(Y(}sB4}h`|doTu%?pxO==)l2rC^9i~Yq;{PyIu}6{!A!+!9@Lojabg<
zhpUm;`8y{$MiJ+(n_x|J-x0kJGkRDn#0Dt8y15F@$5KU_<a6rtbq$+@&nvns@NJV=
zjm`P0up93t8&Fx6=N_a7!RkARD37@;-k$_j^I7MsHxJhT-eLif_9<wZUJjfZ8T+iQ
z=w+Bb_g7#@-Ue1Mi~WgRnD}EwQ6y!6AI@r1;@ATy7Cg1`BK6`Ugg4I^s~FKA$-n_x
zVd5e}7<hy2b&iBnyTY3SQH<S0jK;oy({k#`WW$IPxAvCt2wL6^BOE3P6_R!!wMf2B
z{!jf~Ab#@{T%JD{YYaca)w|U_t*lBU@v`#&24HM`jE=>MH_Gh6K@;EWE{8a0!)A7n
z-#{Vs$5tOB!}CWu=agJ2j1Gl@4ylAW`OKquc28S3$`lG#c5uUf^|GVT)bLn?rJ2C9
zUx*S}mrB7fZoRz2@c0HE@4FxtpnNwhrW5j<sS_B8vvnjE70N&(lqKsi5%$OCv^1{c
zoR^Y3uSgb%)i;D#OFPVlnxzJ6F)_S%pxj?+rV*BUoC&(7=~%i-Qcw1>oMjiD!wZSs
zRZAL+;Uw#pez!+YDSa(I`nHYgU5Hwi<JWiA&}lDn%L{~?X+mzrq(^|~WS|?%1IbWL
z)!;ZFQ}ihA87^+1nm)J&)TxUZyHgo;uqST}RomyvyATtswdPtYpcdS{qd1&N%+7vF
z^E&NFap(cbC>dg|01dG3#qYj;a++dtnpJR`x0bf<<RagDvohdg-_ka_Xp3x6yK8VD
zQwmlc^jab%AI_y9vb2ngBWY8Z<V}z2sl#tZ^ohn5{_784BrMB*F=jkY94r``))9b3
z#%wbkQEAg$l1G1o4|%ufWCTD4;8s*stel&V1lC*x@6i5Q<6NOc14}|O%%HVyk=qtC
zF^O_qBN<AO-#v2blDxo;yEtWDK0xR93GWG--+bAow~-f%>idReIBB0LmsUH>!j0~w
zofSS0@V@O7;77pw*nf#*o3@x2=io#;mVYPDs-I9rALAKB>owTCMP9s*zakr1#Cao}
zY3XqqY$)P_kAXLHPr^$R%Qg$;d=H)>46m$0>OO^b*qq^o6z!4a*`y>oK||IK7GYBg
z*Iv_Nm(L$CgDY`!*fSdPevsDRnmqK<dVoxXe<{qEGZ31~mZn*13s#^yRZ~(Lm@2x!
z!-Ddkc$zPJk#E3tl@79O_gD1A;@c{vf)+4W&y`61&jYggNHDs#%MxVhG~an*UF9gq
z{v`?QG=5XGiGPj#XX7PcL61SWMn7qdC!va#l7Mol6jkoAA8~BA-sVhN{~Oqj7^{_E
z3LK?w@X(RaC*{jh^V`XK%Emjs2Er8<df+yzhZiD^T82$o;wkc$MPo$u#T84Ai>if`
zN;!*TiHunbW5TvAJR6oDdg9`@j*8ahCgK0uP-i(oAoG4tm!D@P@s(&VU?(XFOG^j=
zss@9d!!K}Z_CSyiqpm72p|@}<m3+2(dWln0Wx4jMN$ZF_xdbt>Ea6PIHG9o4@F|n0
zF77O(XsK~gS7I}Yf{_6hcqt9#H6(~Gk_m4Ho<NNp#o$<&cLIi0C1c<JC7}`@QBn9#
zb%>v5>NY?k%E7i)no|$r{lGg4T$H6cPwZ%TtJ?D~hK?WxLu>;n=dP)Wx>vgoJuda3
zOqVyJJdQD03niSuoX@)h=8bhd&vnuB89N&kc5lLMS;}u8jUmG$u<rN52JU^@7b#eh
z29uH#e09@1!tU5lV;d>ubp}?}>G>VqpiF$ct4M!sjtFI4Lg-+&!9EQN5_b+Kd&p-4
zzLl^Zg46_%Vt3Qv59>g7xa}frb(#!Xml4$ab_`?@DBwJAzVO_TgsIs@O%hdW3bDRa
zH5?90IHNRgv13?-DpAZZU<Shbsl0`R$-fac<Dq6Aw@wD0G#I>C-&8eaq`mt%^}d?`
z3`Nc&s3cjr{>Y2WdlukPVROj_&SoTr*CP(x@0>P!Xq=}#g}LnWBdhXL4bq$N%D)};
zfdM6Tg(=f~yh;0GO^Us}SQN<rYNG<Tf-Bb|L{9^HHCSy0u1j9yJ9>yIMPC8kRu`x~
z{5gGX_rNT17TxcmifG*i3UQi;Cet#5ktH_?65xO<4R2OJm|f`N7UfK9RwJo5orTr}
zYacpYI^*dzVp?Qy<i%%l5Vt1$(8t7X9y;&%hNVCAJ3aFN%WeF-m=`qa`H)F-GHr&i
zf#_I(Fq_tmZ9I5<s!gL?*u!$4>gk^Kol{QN0UzBUjeB#^Aq<N1kS$eT>hqzCzR&vT
zvV=;M4G~8I!Ie^Ed}G9v!C`<YsM02UXshu3iO(ALeWu_XD{L~7M!|E0rvp{Ip;|rJ
z_s0_2Hb==m4SCPe$(6r>x5%b{#)Y8RRr&k1_J>B~tdG!=1kB1VM8g%58Ym@`QB7M~
zw9Q2!ygVh9fGXlPmH&t9v#$2y!+XK22Q6332dg$sh}Lp8PKuNUBL?g6RbJy6#SSTN
zpLeEgo@)=Wxek)7`~J^64khXI{gh_fu=;TU%&%phdZl*53LvLa>e^p7eiDW|<c@|=
z3Y#U<X048QSNI?r!{zEX_u8YGy~O79?7Z}%(jhO03i%ofaS?~{@jN4b|Kcb`%_Vm3
zi#9gG$HUwHM1I5@UPffu3NP?eG=<fseyw*09&V4)w_wZ6QT%xB-Opl*ShWiW<e=?R
zlNRh+o1{<y<)Vuf{66c#VTcw)OO89Vi+2;##JKXE{aX5g&9Btb1D;6XL$n=CiZZA3
zxCu(M|H4TqelHk}D9tqBuoTLwKPgKGIXu9kZYI%wyFApqWiQ+nLAc$oTh`uzvux%a
zoKQ1WVarh0_A&&1w?eGMafj+8ZOa3`AQ(8S7&vbe$c!>s|JhKl)#SZ*(UI8#F5A*T
zBBFDv6ZJIkj1H>9a0%G*U+gXMGON*FiRD^J^pJanG^Ka=1?-1ky_X5=KFASvoK5t#
z4Z!F9Yq@wwZD1}6r<!<wXU(}81>0_s_$2Z-@xXUt9u%32HVEZwc&;7w#R~;7z8E`E
z5dKnr{kBrN`dV*za)(kdrf9uH{L*w#lL`9fTiBJ@?kBzx6YNmb#qQ;)g0k?%(M7~u
ze9eAMHls(*6x2Hd5HahTcl@YCL0u^6=p%T;QZPeAzGCzyuXdWXSk_wo>#4*-W}Wnv
zkC}08TTNB<))3m6l6H49z0>&HW{u!dshXzu3$<PC;V*~LSjD(WGlIL9!!CR-W^!;C
zr1E^3ZaW3ni*jaR#s0tPh^HV=NN?hUc!jDnG&GDz@HTGp^K5qRlQq3V#eBR_m(x{q
zOW<FoP0N3{?IAk>09r<sl*C}UN?4fg;U2k1ria*qU>qbaq)Nex%=If1#%9x`_#>sH
zf)Q4uXs46r$1)|b3W6_D)(#e@oDuj0uPT;i-NmFplMhhbg@(!!G<R)Ca7TlTwEWg+
z1?6$he{(-Ek})tK$(ibM=%|wIjNq>*5x7$E3di+YYn>|Y2JgD-@*3{MIH2;6$cQ({
z7RY&@{Ap*zv#WN>t?Z&@5iUIiZ}9oj)$@3Q6};P;3r9Zbjbv94n4?&J{(ePWE7vXY
zG~4a&b@#Vh&vEgKk@yae4~L*2Gdxj+7&ZtuJPm~axWdQ094Syv{DuxZdOp*6NFZ$Q
z>%|>$Y~-y4+{=6h{hZ50D4Rcb9~=`K^;30Fukg29qe5$~yDy{r3oBshl`7*C34?*C
z2Bh0sJsYCbn_UKmzw)mLryQ(=UcE|JQkIj}?F>esotd&So_tll4}POsAyaHJ4Y2sF
zj%mmKavqLoWSv(m^R1)<f-8Rjr;Co>;l`E3b9nM*)!_>wy%s(Tsk=F``hCVuFZ#(G
z2v&+8nEz<nc_y2T0ljgMT?oK4zwm`N?(@-I{fm+euoOjoe=kV}T*<Y`ij~-lwY)D)
zKn#&-wX}C&`mg&Q^r&x>h#KcT25AJ}i^gj?I?eM1Qdr%24<J4)a+^r<EUn^U?G$5i
z@MLlE7Na7^*{^?jIAEG7kE_BmSiyh+FLI`RBAX37D96m=fD4eEuyYn4NZm4*6Dh(2
z=ni^$I^zL87aE~k4UhVu1WILAcbv2v(2HVdWEGA1FoF-miW7Fhs53`8R_a#dDV8y{
zAvTxETUu@`eS>>fH^w2#;9OXxf}zf@o^6tE?8kFAPx=fKXLAnUx}&NrvCOwN&igd~
z%2{a*0f;YTNJ~s6Lq>01t8;t=pSIsSRge4Le3C&eT`uLi=l@isjK0L9Mn<lJH=heg
z3e_l<eGM2(+7gVM51}Mkxjbb4a$xiEqtV&3%)&`uNyhWVd_(?KR!*h65(VMLmkX(G
z;p8@V&C`Mnwe^Zz&?ooG#dsQ3))iv$i+dbDm6nGl%7tY&8TzPCwD>ql28At1geH7K
zxmDfSQbJ8jouGqDfBqB7OWBv7WEJRVC(<VihqPgx?+2p_zI9}wJoz3880se+#}+Mf
zl{DQ?)HQz4>?hj+*?KGhnHwUdFjEst{hWe>hXv0{@)<8!*!3Fz)#o26cU_>}iA5bL
zr6m#7-W|0p7dx5`zt!zSr*#@kO(;_ga&cz$y+7(|3&ImnE$6yVJw*inUIQI$UM9J;
zo&6fA&O2m-ZRu0}so$1yi0^>LJRd_F(NUo5yKr30rzh&$se=fvfIE@50Yih_(XOc{
zm<mG#Ew!@*U{{#~eBqRIBm()YOGvKrSTNh@cPn~U`Yk&7<xxXQoTa^6Ac&}>ETGJm
zm%!tQS}ua=D=fe{V~<c%ql65)(h1<m!hg#8bi~q^xdPgtf~-R>QR~@O!0qgK@^EmF
z0#b-FgXr{2K&Ej%y#)<L1z0|Pp4`=1=g?U!BS}i_{-ntXI{EWe$6_Lt!bo7c^{c%Q
zHRxIhjQCuEY&-$%UOKf>dR(Xz>A|2%6V~1p7WZ>^bhVN~ADQ<=)LvL;F<}HfKj+<V
z+FJGQ-)GR9Dy-$2TzUA$(&Y1|`if#Pq_V||M^tNL%I_-^jd8un*rR}CQv|NGMKhXh
zh&N^<e7jtbejyD&TjJ4ch-L$O_81(TD8;<TnV2P*Sh}#{|KULy0v(RDvBYOy%42_W
z>$dk9E;>l%eS3H97R-n=P6(V2@-nbxN3MB+=;5+mkV(rd%=j$Q4YfI=daD;oZKz}%
z2ru2)w=jq_S4E&ZC0d<7+H)n6fS2oOpZfr{xM{U@NS;W4&K5twe=rcO#&YmK=O;aA
z^e#_j|5?*zT?qWKxYxjjhUfxB8kZhF?u0heI)P11D;*s1C7*<>jVGJ;?+Aeu=Ggq7
zP{Z^ITG7g*yW}(BeCC%N{R>@Y%~x<t)X6<J+xic=&^$^eF0>4i-rJ-UCx(%$ncX4}
z9amFuJynDLXZt?=C~)!|89Fs0t9+;rTj-vDhq(ly5v$1<Q%iNK+0k@PTFV**9DHHq
zRUDwj%u(h~8&v%1w}c9aP|=4$Q;-XW_01A5a%Ths3su3UWV;#!g3v{9_tnCRpa=RN
zs7cZbXUbYrkfjC>Xa2C;{=;gln!`$?!Ai?G2JiPnJw#~(`TT8rCe)t<BuQq;mb3=}
z<NnT34EYQQKHY~<Ki186o?Ho`haXE;<zM8Ix!_1599LVf`0gRxwRT1R=<~s`=ZX$e
z`_d{PkALqxCNoEDp<t1SMtaLsN9Ty=5AI)AK_BwICz|V~>kvp!{cw_s`4Ko{9m1d2
z5?*8jMH=1&92%wcDv_dqUG^Xw;hd3Ub#7i^=vmX<2LUJ-0a&a{kwtJW)P{u4x@7Uj
z-#YF^-6TCmiZK-diFDrG)(yzs(hB7^@VJ_b?GNl&Gz8jV^{l@Xus}am=!YDv<0XEj
z`+dn8XMHS)quCr@V46)|@A$Z&$ypb6KlhX;Ay0;Cd|&x~xV9HIX&37tqE*a=X)3;x
zMtf9Jx4C{8fbs09d>w$d8&_)psxsQmgc^L!aG3KZ5D90el-a(ny30-_mVKL=RHq7Y
zv(&9KB?{L55QfUD+X0S~dsQz`<e~ji|MqSov8^RenYm1n(J|GinEMLKF&($-7o;M%
z-6MSIz^=yTPS?WfZsD$$_Hx~eh%(j!89fTdfG{TUy)bt=FCI<{q}FB%q>pXjL#tW+
zAw9bHM=|7i7=*<sW(YPIZ6r~}q-!lD{4Uj<DD<T{zMpUdEUuQ=oK;Zo8R)Z1EQ3E>
znvCz)#wt{aLVG+DS$?;9L;9C*`6;)Kzx3%=0$yr*G?-HGcrPIRaB73!OyTl)KmYLi
zrc6*S9L2nBjQTh&Bvl+U2}P@|jn*&YN>0O~NB;4$0>e*w5<SkW+HkVOf=A%AOp_9;
zjp8QTZ3m7&jacmnEpqmPm>t4;i9>YF$*Z<zsoXukK=I3Et5Lb<-7uDOB|bxA6kz!q
zd5+?-orfOA*Ckz~^iuk9A0@lkp%&fn<!6k?VFqw703q7V%6jkU3q>+KeNjn;4cjat
z$d<<aOrLK=>&F6-yD6=@*%Pf@*kL{;fm^A9fcKL>PJJyYa%D>ce1Ow#d0Rmaj|YBY
ze}A`^gHfF*_Y1mw2W@c;7b5(egK`9KSC2d0Xd6B?xvB>_jQN>;EqHKN!<isA-vY{c
z2s;P27m*1>1B6qG5;H+m3QaHZnodc7`GiXrk_i4S)!B$28rVFZ!G-ljr&S%0(3(z$
ziVJB*I&@BFMcB-XZs9>yvk4n_`2%;fup$RGd_jhxR+_0I2Yj3-VBQKw6pdq`ISaS<
zy(8gtNDx(Oy4t638{G+yH`^nhM{q7LJTS3h9*5QRq#A9KAH*`oEdciuh>6pA2bf^v
z3eI(3FvofaP(W9wX!5#TVNFEKx=hyXAxP>J0D@O1!+6a9jSUcdql({`&@VLYzIN#D
zVw$WmHTFO5(EDh7#r&~l{duIx<foG?Q(T%vEsrm@<jmTTh8Y9BOmqD@`1k|y_EVgV
zm}N@dMZFVh;V6z4xm71}_<QeA4cJKHG4UP6H$uUGhSv!iV2K9PtGBgFB{rCsI)iWR
z5{08XYZi4~6Q9tEo5XoLfYsfOI(g<%!aw43t&NkBm*q0b1m$Vd0(a@j0Y!d})WCrt
z+IC}a(z?Y;h6g#*Taw})n=R5p=3%Wx6<VNY{Lis|U=)6HK*rx>w5HsmjKzypze<_N
zMm@q?Bk8!0_EQ`lIq~1STGQoFMjhkI4>=Hu8(ux(nlCyzSyB{DO;odFjM{;-L%^DF
z4#bL?t|3uisvBRvUZWXWRZ{lU7V2i1aoSU@{^57}W-ftXfpA!=pr__=2^$W?)z}9C
zb)aJHx8}_W);Ww5!GxRHrNpBEF+%MKhw<oS1v=m0^ucP;c=0|Od(97ioiz3#?z0wu
zr6{APW7YHe*bJX_ws}!Jy!I}aWAy3m*9}$W6G32FEf1!bHac&9VGe5)-B%hi7n=eK
z_M{H9xwWt3(m4R>`|}<Q30L}|dmOy7>NY?~5GT3muXY<(Q89K6=*8~OJ%5w%4vvdu
zX;N^q@n?Ehk!M{^pM?+}9>KV07Hmtkin?{Oj=c;sHNkOs9U6GB6ntd$9AHzCp6Ym&
zKq;O1w6MzP@<dV>mYjJ6+)8L1MxFD|w1FuFQYIuH2|ks7M=kFYuMFJgSKK}3jhoF7
zC{6%61ad7jH<?{h)zuXLY*r*C6>edWqEYE8l2TwAw2qus%?e3Tv{yt5^*te9!=ssf
z)E004?=;^}XdpFTeZ9)$k;0c*0MhG(FBvIVgFApW9ma~L_R+lG!>c&ywKP1tKfV1B
zlV|9n@dmDWACjsR@VT3c@>B#EhiMoQ)S6Ht!E1RLB6;fmLkc|1vd-~-CWC?RlE=eK
z4jx<yi#ytUaSTtDX0W`exTdh<v#IJG<>B`3<X6?WdZ-4Kg~ywhpLXyH<0?#AXMz@%
z0hwB#{3-ASJ5+XfadkE-aUhdS;l!c?9~R7Q!S$8oW}51LR?{aN!lX_cU=~W5h0*sU
z4}^jW-3tA2w=5*9v~-K_!^iykSga7ao#(Z|p=R%&1k$6^up&Z1gZC*FRD8opp9(4@
z(3v<3C#2k88LWyS+c?GJMe&GGLk1I#MEa+|oA(o}IT(Yl+X@v`fdk2EJIM}WqkQ6q
z6sebu!#u?cpRh$Z`*LUEU)R&52otK7LZXw)Kbl@{Fo%!Zkatl50Xmk33CCf^m+a`q
z>92&PR)Kt9lCKio&hy00Dt43@3%?o(HBsQgMJ|?2dA>>JQ}Wt8V>LDo|Ew|)n5^!=
z-sP5uj}du~b}DG3kDKKPzlXlDtJocmj60|!Xq@Fc7eAKPB-r}l!HcNvm)RZ(RkB9~
zkK4@epVck5iITwmeuMCfN0Xg0D&~)P-Bc?i56fKjmBpgh>6M3Gj#W+hYbLc^jxmS(
zWM4k4i&}Hu2!FuJ5ng}d?k>YUiQm3#>R#Rhnz9oOM#3bfktS!LtTMPOm=yIVR3{}*
zx5&uW`{BN{hw~{S%Pg()oKB~V)`4RcuDsoccVeS9_4Km!b9nWPTk!MOX}w+8b5}b~
z;tG1Ijt9gOtcd(=55~ld-;+QOHMf~_r>baC<?`AE7Rdg_lN4z|&F>z!^RQ(CSdYV8
zZ_$}+=*nw4U=E}fjw08zj;Y#%rJQpg>{s;YcWnV{0lQ`Z+3kAKWEJ?U{mZdxS74Mt
z=oP#R#AYrFIq=aK{H);fLjt*z@T&MUZf+H-W)&dnqJD3(xkHoMEoU?GDG!7kwfDu!
zT7==1esW$2t0jT$n{xq?#Xnj3PDc&=v`#1<X%{lpuv7vn0^~^sY~C!xQKiyocv{GK
z*1bxW3cM#^UzFtmei9alne#3rF5!AuB1={&`yzDeD2SigYFoH`x}A5a;-0EQ*i|X*
z)#&xKrIn(A?u3$*180M1wW1jqX}F4LGlp%qr{Qu~tPUK`Ip=FJCZ;3Eb7&?C@5c{z
za&;0cwtb&3+~lnN1?)S<K~8-5t?CYFj6JXQnrVC#zJ9rGRBj~`p5!rc6IfgGRA}R+
z9GjdYtTV)BHC2w2;sNQpC_2z(cA|9*IGTH_<wqc?xCqDHqLCs9hFz0m%Z*$;kq_F(
zq>d8^4mk||AY`q%{wk`}VhKs=5}i<h{i9XTK`KUE!!tWxk&}RxOMf>2>KFg~TwhS9
z9fXdn-Vwt`+q7Q<b9MUA1y=C&QHZ1gk3W~hxu-Fa4d8l%9xm7w8Q^!N8+Ts)Qc?aN
zlS=~Q+L6+dt!WZp;SrF>hm`1Q|F|-#WjFVD!_Zvg@rN7K?pG;G%<cYG1p`(}MmkjM
z)rUS}Qq=w4{!#7ge?=*xc6ek#{`B4)c~MR3Qp;{Q)6-aX71Zrl+1GmC`AlI5jZ3)(
zDio%~K~-WB!K^PO8bI@G|BqkmPj!VW`uT*#oM3g!&3G3o>4d$r=o9jqvgnBWDv$jj
z-YO)dyHUcwohm`5c`0fs{vS8Z9e2IkIC~|+arcFy<ZlPFOx)0S-X_mtwKUJ7oi}QX
zBhp+WJn;3A@`!Jo#1cc%p!C(9Vl!{vU~rgM{PnQCdvz|cDEq~s(Gy4Jp}?zYlk<;M
zygNtPXQh@p9jxPJ(`N+uvZ$%RrM&V=6O5&thLOXgV-`c{jCRX;AJi}j`AN9j*2gdt
zk*vuFGt-9>r@D?Ke@vU@Pm7gj-#mG`(NKS7^AvYdoqi9GamMD2HymHkn`-#)#)qP_
z{^t-A*@yH4)7gB#Q>lW|6G_A@T9vjypFk5HbG+x0IHgg%zwA)67DwB`q{aK%s8K%W
zQas!#rt(7~2l>rR?JFj4-32ZQMoXJ_TQ^tb%WcMP6Ic{sJR?CogWgQzRxJR=NO@}u
zek}sha%C-j$rfzC^cq>+^vkj)Ld-=Y^r=1zblLE6SaIc#ZCds}YkbDSyELKiJ(p=c
zrsI7hnl+LPju+p66Sbk36#6=aPfg>grh<s=>l-LyfeOuTN=o{%@)9E99-cd`dlyre
znZ1|Gy~X-LV=>Mya7Qqlc(@1t+43ED3+6XstLK_uv{lQ1@|PU5)W)w_kb#0<5zdyV
z9sUbys3l~al+=*!HiuCUta2)pguGH@aG1v>7mhu;$jgm)>$LuOBnuDFc(5X@57U0I
z$_4B{mWC!ZzmqK|7p~4acG`HAB=@(cz34aJE+omOHtZ_X8+M(IeR6X<BpD>T9Xz)G
zs^ibw#kX8Y^*=&z3B7Q$)bz(yQ1e{5kUCQc4h|cek$lZ`mdx-{FEpaxg83EXXjGLh
zMTc}hv$e?rbre=H9uO1N01)Qm!M-HOTLm8H5lu*$uCENqr`g_#^fe~iPr$dZ0^Y5$
zv0uPjZBc|fgs2%WhMh&6^A~A!UUAboAU$x8Y^W0mh79Ra%T(ICem6!u47AIbDccS^
z^Q?-;5%{*zhS-&;!m_VnUCDJ_iK+{)D>&!mlI_FKh6A|Mm(wp)&gft2QFn7|kz}TZ
zab?48hYRPo=;s71l>O;gPjWn-ggQdChz$}Zcw*=RV^nBMRZuDK<5;)}af%@6=8a~e
zRv$(fWy>yUwNGyn2dT-&q$0IFB}5d>Gg4#3Gm5t3{P+nVbYP<c9<ZyaJkB+v<7F8k
z>9Jt>fdEdvIvSW;5gMl~pcKUXiO+i>!Ijv+4Dm+(Gbhp-@#of$>Pr+ZB2x{el%7GJ
zxSgk4|3}tYhDG&#fm(V{L_mpQ=q^D98M?bcP^4o9PzOXDLOLas7&?ZMMjE80m1dA`
z5kWd624oPp$KU_nPxlkwc%C!+?0xoH>s{{}Tk*<XF_S8~J)fu7It=lj_6p0El~mrD
zr3#Tu2@*n<=7x$s^wL1wGoxPF@P%69GOVga?8?%)sm0(fmr34vu9;URosPF@BX-mp
z2ic<<HWb4>lTUROFrpil=z5ng)ii=#rR7mVEX5&DYc*=EYvnUDMzaTcU(7`M2`=M)
zJfmD^?Nd8LG^@`<+upYWLGw6Kf<doD8fKmLDp6&GV-p>Wz<0<x`(0*%d&iyA8k=d#
zB4-|8It<;pNi8Y<KJw$ziSSi~!t2<Bd6-6Y@%j4I)hEVrWf9wxMyX)kIsF1GF1ruA
zeqeL9BS@1p1?TKu&(OBk_$@T5+Ig<CTpzVq)NA!X+11j(ac#sVzH8Ahgjfp!$<til
zhg0+qH{&5yd=tHXyp4#Ta=rp9wYN9kR@=2sfymNIv*hy2v7oeiA+iwVT<^VqA+B#6
zDNVZ3ykh0)+}9Vm_09Vq%ejrX#-tw#S!?x#CF~^z8(~uCGB_cqIVEhsrv}HHNr~h9
z!7)4`uqg%FN7qRs?X#)XW~+&N-+Ln$x48zFyHzP#<URXeKEU?|+dV9mGLQG>-QcO)
ztQ6_YCR}>Riw+Km)qfSv(Gwu7_);ChLRHvYBd)R7x^0zvfBuvE_sueSuVOYf^!=iI
z_ddesYhe_=<i$h;?U$#0S@nvU`3WE1)kyRjSB;y6_>)`syOlR^sD<R02iyB@YR%j)
z5O#n;i;M?IqY0rP?;_W%%!M~KE?Wc<7BeI6&ot!gn&})l3-DZ_DX0(;Voi@A=HJp3
zGqab)f4M^U8f$mj<E;H;2*G32tNoR1hA7y5HhNT30=;zBfqR%<!g<yN_3J`oege*U
z{+L0ov6(AO;cQ<(&CVD}f!O<NhiT%G`VQOTdeIV{_rq0)oOo<$E_45wxdyjBh6H_U
ztysY+jEe4?V~57kQl0e-d*{|b2(_I`h;+@)Zy_D6{vzxLqtZZP7`am5wsB5N%h8~q
z#MQ`nN@>u0p*s20fGLV{QL3jgo*lRLk7WYnR41SAUHHTuE6?(+J*&9!vl|@wXMNCz
zHZ1zaoD>a-Jh2_h2HVCW#+RKRW*B#H)0y={WQa*OEF?;`#vwviTYcyNEU0cwnxCS?
zFjaYyiCq3&b35s-H>GIh&VpjAWNASa__I~sFKS@Z_HyyRxU51B)OP0Rs^b}684{A;
zU1kKwiV)xD?zn<H0LQO!lg7DW7L0V-vB-^nkhH=2WfMG0G1#{$)i)MPhD9!HTE3Ov
zE|ZVLCQ_S;7SvYP8Zkj%i)qY9M|lF1hj}~>rE0UTz^nPg9scB)fEnC}6v{VlE+Z*s
z3Vi6quLGiMYy>4&za(GpG4)t&RF2l>YNmTnh4!|j1>78qFCDj@50;(x5Rf#Z@f+ka
zziFHJ*4sX|Mu#wR;id11<M-fm;{nV1UkIf@0qS~H;^2`p{@L_-i6e%b7L?Wb6I$gP
za4aW})d9}tfyNcoGj?ix(`dS858lzge64A$9l;;E&h*=r<}Ng-vJ8f;DNiLHa^&!8
zLEdGucXW>aP@Xj6mZIT26#JNAMu^}8X5UVC_%>eD3H}$>PAF}M!Z!W#jqpX5IhmRN
znoDODpg?u(y<+uGmE(alyf|r>gqF)#ozG_^eWRo$J(7SWt4@eLj^iqhP8Y>CTQ=Up
z-m)X2))yb7Q2$QEDXrYHPO5-jsquUV=YAFc-6Z_EX{0Abrjl(WnCHxzQ46ayxqT(?
z%ITcRO0&>tmB?yl#>*)%sjiB>)?`*2_EA(H39IN>4hICJ@k1@`_h%R^9Cp#%SkLDT
z&%^vFSVHe5WLxO~G<%+OIAPSQ@|}DT>xYS#cd?Z`vHEmP(uC<`eMfDfHJ<_y?!tzS
zio~g#^$D$R<uZMQVTJA38`OJaf`9HiH6cE<!=5iph8!ufmccjK@-qtARSH4j>ANwB
zXJDL=4+t}w$tzb0?kVp$nv;w6#_j=|8uIGi0CD=AwJ8_KqmRx=>x@<0-T^1)Nc*c{
z=hSGfi-elOI(O(yi<ii;0fBes;63t>lI-9St1{w2FK4g&7VHk<JDA~2(T(8~NIYC7
z>8-;<fppaO%XMf3X!scbZ+Z$WJKR(ut?%)3ycHiMG}ABn&3kA0$SL&^qXVxXc%Gzn
zUS!YIbDEBz2{tW`oF8mtH=9q@CR-{$jnpf7OqFhfT0T%><N7_|t@<sk`ObJ<(x)2k
z^|0k8r;#S|-T0+ahliTAL7b;mGz}*msdPkV59wt&`_Kk^({$}o@>9#Y>>XoX(2|0u
zGK*weGZM(8tyk+(;F+MYzm8R}u?%2AIGtR6PGolbR=(5I?J<`hG?^0J;qlX{=L1ZS
zH@M#I9Qbch=|yps9jaIUy>CPmPu?QJ*H%tP6NzRB%U88wjwJq-!|9Cgbu+FUJExYk
zDs*W5Dk;%HPhk8oJRuPjoUX#{OmGzW)cA8IN}-oQ>1Ro7t#|-yH;=JmYCqp|xA?dK
z-aPWn{2CMJ&mbw9^d9~03W*1YM{Vt|+-`EfN^xUStNq{u8r@fpnZm0yoMf4fbnQeD
z#xtr-)2@FT93>jq_b?nPbJQ}@^k^eO5w;!y37?2><h8?A-14NgG_Y?k31VoT;f0Ap
zzJQVNY@}D05eTBfDkWH)&N?t?!sc=E%^oj3`r=xeI9pqGSI25?H2n*;;0Tww4dE~S
zmY-ta*NnCL^IysLTK&2fTMm062UrSW@2rSzVluu}ivcEP+!@y#fBI8OgI4ABzbX0o
zsm9pP3G?y5m0>jBRWooTEl5=TBpL!O`)n0;Dm*(CklyO96;9Swg^0O+gjZR=OGSpi
zXF;}`b_7v2aO#h@R3}IiFUM1WB~I<hpC4PMznNj=R4vV(aZKkTK)8y5K0m6g2pcii
zBhYE1TMU(5th+#kJ@z!P_{8sNPsxsycM4Fknt}Xwbq2Gr-PQjdLaHhWtEHJ%#Kff6
zOgQe@3(g9`oy%jq>sNH#zJo9W#jEmSFY;eA|HAl)wRmtrW%7M7e<^aI)(&erO8p?#
z2!l;2eWg6xh>%Zz7lrxX<w*Y#{y`-M#*uwcj&#V$9O5BEBoU%i+YsU$&cXee)U+|L
z`2wm$uweRL*N3o$>N|G%aLwSWS0E<od`qz=d9i!b5WWSsn+yHAOJ6!?XHcAIvF5>V
zI8^OT9-*b(jfO5^u2{aEQH0mF{|PVY7xDHBJ*>!E-xv7}rzm`Z>(aAWLx*~ojgOlq
z;azR_C2S=H?H<OFkO9<ubd7Eo(|LLiu*f0y`hSMmIfOZWQEN%Kx@bO7_^DYmFJx?|
zzbzq0y9Bw_@E+T{{%tEmZI0B(KK5on3=vt^{_8az?K0Gl{$Ug}!(4!h%`x&kNhQW|
zSgWPDBcNxC{NpBxO)$gF69@0@McQ2m#KGtF@F6wg1(j+?X_)d>()~5f80M_94vTkI
zd-M@iH@iKb)TTHmdyC?;kePRxFs`0M9LJA^*L6p0kLvezBn0e2xeG?br%Um2M>04@
z_FCFu5M#^8gCaarcdw(&ld3}qYFMr)#fI&w_RLT3ew_Y?1o8gA)AZ0SEsEM5WB^X5
zcNz!Ew89!cCSP5ZhHIUTfV&j&3LL%*ufBsIk;Lv#cy)UNvL8XeJuVG;!~r)Bc_#*6
zd|+>6{9=4Tv1z&pA=*YBA9*&Ra4EJg9L03=-f;%965J5SUS)yzB1r%G#tt`iXkw^o
zB&qg<j?48IANAYRm3M<1nGd}k{4q`shI9Tz*;L1NUT<#WcWi2W7FVzO=cz<^M|5yI
zBG<_s_1QuqHvWJ60uOyYFSm%k$!r@y)2DNQwb|1x*M3|Vsy-sUH;_EFqN*|D{>Ic?
zLViA(r0aqTRxSLFp#Svnga~?=h0?Q9slPQ2bD3KA(a%X~*UI^q`vYJ-y@&ud8V=9B
zZ@kovQkhE?gJmc4Pdom6b)AcrWBzk^M`YQ}<6bvuJayJlNAm|4kNX%3`;SBeLhRaE
z)uKN^gM64(pJ1{)`v7nKnKG$Yp!H9M_tsG!>LDpDFJH2<$?qGm?Ec(jJjEmq(!Aw2
zX=Yg1?IeY>>6BfIi$x4Kr|8NqGf0Lfy}b2BHcpsRBlGoO&a@%B!pZe{;1A;&;&Wys
z<_~si^wQwCYM|ulN1L_lt2=(rTtqqoR`U^QpELq-T6Ez7CUZ|^90$lcN|&62-jOqB
znV%{)zED&uuH~FQkQLeJOw*5V$P~+V_ak31JcOFQ_^F0=3<#?}tf^<(H*d(TB*TQj
zs&B1Lk(+(pw9HyYD8r;Q9F5JM>_I8tytlo}FHT3Jn;=*^%^(9Wap^Vbpf(gChOPHz
zQ|?0PM+NP~HbGI?X}r%{8YaSIoCPskg$U!;4Cxa8AWamX-rq6hmuOMt-2~A=H&p@T
zT<!Yqx9Z>2A2a)TaQi-c!u>fPX$VH||JnZyiNC8NGJh|N^Ywn{)AwuJZ=wt(&By-U
z&H3Y9rXzBxV47N9zajkhYI{(y!=>WW(A85s%_Tmg>b!F5vFT+e{(DEGcQ1ewtZ#=y
z<h_i|c_bW2r4`SF1S-Q&`m8cXdFHLiRGI)q2C;Kr$B#Sf!c;=9cz9$~?!PCM-(XhK
zAm<xzKvTJ#S=lIur2dy3PDzl3Kgvq<7HLI%Ck!3>z#b&AeDrq5?Rq3`d*+}5w{wXk
zt1kci%A~u9tK2{wWd{yQ$QX@$BC^b0HSN!D@hj~ZD!ov<v7X15ef!~^#&~Ma1mS9C
zx5?TNvpan6xgo}Zb$8IBqrZLVFZX^id#2>CPi`MlZ#P+Vr0%9TRSy`$WNfSIyY!zV
z$D^fwSR|z1Rctc#&&K7rPtE@356MzEiLh>M!EYa!?@V_rb#63vT)x8NZXe*kPjRxI
zf3IJ{&rMw(TyySpuHHt{oMz~uop;A=P|7*8wqL^bId2`3&+8v#Vvg=Q*PW}xAzZgC
zu`wV>w5Ej|&ycF4j>PpdOptZBB$5X=b;&j)lqu%uk6K`wCElT`+n>0U7KJc}%69;{
z-~M;Q{gu|Nk_PLrdy~64%>xpH<ub0R&pekIjp1d>Hu#L0%IwomA`8)%zyd9r(|@ny
z@po$P^uM{0Tn)ZQ%?5oS&H4bHDXO7t)Sw0ot~DYhj1mnB+h^zt!ZGG%a8w!UKa}*B
zoBWXLI!(l;NlFif8a0A9Tnx_|9fRr^vX;ZGX=v*c309r+o>kSr+2RW%+7FRW&YDLJ
z{3heGqbyu6L&PE{?lo+so-V%XVP%zFRVdFsDWG?m=P>8!8$0k~cka{@=8y1vpr5Di
z^LU6<gWWk;?Hb`~*|pIeb3?~yp=bo4`us;mY0z^Gub-*$I@9*wR*B?V^4=(R?QMxu
z$V$737l_rKO}g5WCp3&3;!5v{KH^JD5^p(Av49=jlLa^8H+I(9ka{qX5Y3wJolb;}
zq~Q!MD~bJKf2Lb{!f|CTJ=0w$uGLPr+ASz)<n$s>$h+?>W^Vot9f&_W&HVkg(NMIr
z{^&4&6bI=%3zYYpafol%E$b7bTF;^YD}S7q$lxcJiK>p{M1_v_RNK1m*Nvv5-dLmr
zyoA{uG`6;Qo-L&igqZwcM~tlgBfDx$Gz##Z$+<D0Oi!E?06Kh{O)DzEj-Csp|72nn
z!|&ZArkH;e_c-K<g4fBp9t^Q2;cYc~um5wxGBq*FzFLQs$nWIzfYUmgM)2`RB2$~o
z;3<RaVd!NBzOj8|?wV3-HfTk=`57fsf+lL99;HuhR9BWzYuiog`@&Smw)6f^z*$3F
zyLF>E;cNbvdZB6&Ra-h>9%tfnN)&Qdpq$m1(SH}N;9cQ&JqmuuU<pV*mY9Ey+F=6x
zXT1DvnUnuc{sv2Y|09JtIbkZNPx`=^1?j?kuYfaUVTe|bBQ<30SNV1hfnFFbC%<KS
z!bri&CVQeBjhWqd-(fB@Wd5ansX@d2FWp}f16|hY8v-oXe^PJvU5)DmzRrhmW|jWS
z*WcK!ytg^Eni((D?5rtgZ8s21Q{3SpTb~Ih-ud`h47jAflf0$FLS=S_^Ny`HKeg?H
za0Kvm7kyfF7&sTk+z(33sz{G;+1yjY_m?_yH7cbz>YvhjXFI#|qH+$wRc1RA9a1_S
zi23%lUtSJI8vrIFe}?L?_piF1-)}PqeE6m=y+#UXDx_9dZIU00()!~YqSB`9WrH)*
za2nc>1D|rjfokj9>DwE_1P+S<e&k6YZhhhhu#DebApGXM3;Oe+^<;o0rrwGb->kSi
zkjui!S^Ke1{YUZ0*;sRKEc4aRDXrc=^T_4b(|;zSMlU{}-?n=7og7kI8&ewMfiUnM
zG^T4qI#|$!-nf0UFny?;gs$MVV?HWE5x5LH1iR3|SP7Y4+DJCvTL57^WHO^7pN6j#
zVRG7@$U1~oMwd;A&wO6pujfDk=zzjWk3vJsOk058pXfgMQ*Ov3$%tJ-2(WI%Pj0eg
zVM536_6ZGs?5oiDxO^5&lj%Tk)sQ8%QU}jRdMUYL8g@e4Cs%b89qBe=u`#x~(IcCd
zUx6>g*Pfb=v$pH)maGMd4WLvSY-)T$QF&l3^xkB8l{PZEJYQUM;Uh8apxjlaiCixM
zn|Nf+@OZ-viiylFaiECxJV3MC(aL=3viY_Sl%<Qw&UfrRLLoM;+HH`t+jFnp^}8fW
ze`z*0^9rOHT5p_>qK)8j24E=JlT{tNtQ{iOKb^P#oQ#R<<Z=l;*zJqMR1Y|oQO06P
z=$Y+QxC;JwI9FQoEN~^D8DTd-lbbEJ@2a)DrHjimJP1AZ%&jCXyE<PxwiY2%frfBp
zUF2JaA1;Aw0py5w1h<zPB`v!%MF~K?Euim)bUqNQdK$Xfbi91Xyxa!lUKAt@d#@XO
zwv7+WQ)_Q8oD4CB_Q|g6ImxZcpPg%zIVoJ!NpL9*TA2wr9K3TOP>qn^HwTbzPS#O&
z*15jS1v*{lP-vpBpk6V>S{+<R$I!UMWx^5I9cPuu+7w&UW^ZhyCW8w6xo*rJ&}U|b
ze0|!HEq)c}zl)J*84Fx?GwBA3aUedcsnu!yc<}gx!E7_9NcXQ;{<D$2%<|vDqN9@v
zNV+*e9V+`OG-s^fPBs1@u8iwrX5>S>Uw=XgG(pcs)5O}j^|rg$Ckv06gw0zvf6~j!
zs;;ngCqp$I7iatDtAC9Dy*#*yj_UTAv&M_-RPjUq%cAjoXLS}uO0!W35{E69^sg$Z
zs#}odYWLd~Z#OWUejx8-UiWdJpi-#VWU{m^qg{_{(#gZOyOe6)kLs4=9-FD``<$w}
z+%etdKwDB8xU-OXd4LX!soq_L`xRL)2N$p-`KdF4fZ@;ljPAuHe!*@-MS2d=Xv%`}
z#-3va?&p>z8<ZqrF-ob0M^-uD%`2zpo^ScWyY3#wV&fE_&<QT6vXzy{?wrks>khtl
zx7QgPSmX-WKelf0`5mgz^rX8LMjWZ@>OEOz+VMB8!D+@eD!~5_r$So`?n(T%NW~$f
zEbgM|PI=aflZ&SClC7WKBbU6uLG|aynev{mKI2urJm&`TCxusk)eXNdx&Pk2_2qsD
z+B<f^UyASdoalz<9MiY>hUaCl)~W-1o&j!N3Qe0HzcfB>1Ek6Z)6OgPddJfkEr5Wo
zswrgRB1SyG&11pErkO|Q?DN_iN}x)ehk$WlC^}_-{q<iT#xwZhf^3B@Zw-5QU~WAL
zUU6zo*kdB$u=Qf(=qH$#_zNLBp59P#k-l;1cd@WsL;lP`a9fJtP#!}mw5sVCbH%>3
zU~PiE0<FmzJcvzVN35$XyR&WXkM3Y7uYR)zEX@Z8oS8Qw=C4M!0nZ5vhxPhi7y61)
z=VS-6e0c?+(g8z*MWMh?!us7=`pr~Xhx^^6rvZwd_?I#Hx^&E<*TYdQ*&~@=maIFM
zg0b+4;G;sTwXt*VZ@dUlbFJfuYy7is0ezV|*BWOGUH+SCF*n=eK=Ydn|9#w4pCDlM
zTQ_7=4a!}_VZHrnlCHbaA8OpQ+QD#$ZuB`3i|s_$=u7t%Oos#BAnI|ZX*W+D&;1y#
zXgWSINhS^dc&<0W_g_CY^H8JU>4FFqj&g$q!;Q6wlystk-PHN?+>KsbkFfvN$~n=3
zI3a9N(GwV-(Zxq>A&Aa5Q>m$h%@&zL_D<h$dOUAe>`w~aDq_+EqZlOy!5l(A4Sl>1
z-`h0%{Jg7x+<U6vyHq8zP%2w%(y@39{4`F}c$`O>!BrCd%YJ?*F68?!h^6eWbI|Dd
z64(A~%Nzko^SiNbKTCiZDXluM*Jt?nQ<S`yQ#)~Vq=}EmyOTL{-IA6)#u>a`?ESY5
z&+{8R!CQ&#Q*dLpf9(rUCaZ}R@7~^j8K0E5xZX#Y=n7HTbl~qqn`-cdSKvfW)+ZuP
z;M9GN@n(FaWbJIaL~QGlij%7}u1sJ@eZC8`U6^g!?XR%iNXkmGZTpGX687*MhLW5m
ze_iF;K3DlX9craKIm<x&+&$(B^<n|~_pz|%6RVCP_2vk9obH@Dz~H@8pWV&-L|UGZ
zK`raLfRwxHynZzb{P{lWI$BPmgM0s1;dd9S)lTo(%XuZ6Xpt;Vq5*;y+Bvq0gqtV)
zT$4J5)Tm=S0T~abD%>~yE0rhS3&Owu*t+OLVyNwzx*hA(ri%pUb~={-xU`u@xWR2U
zNx}G$$h5A1AyMn+X%$Ga+lbcLLpprpUOh{8YRzd!O)=3z3gR)4^x_eyxGP86Yz?O>
zt~CeI+Ka!_mtxT`KLA<XB>d9AU$jeO|5NhA`+hu5gTU#yIzrBwkCQY^tVH!CKcB|;
zp57I=pA?K5p-k^A)%xg~1gxU7gZpFp^{iKoB%BRx`}p|s<^t>PwQkd=J@caeO7A*f
zcAKYG%cw3`TV+1Hv*n5J2Hgv>@B3<jDOC4cjKqv-BS|&t1nu&PdWU5Ew%z!>M?<YM
zUR40kCf9y@z9?pXeiT^W`(uG;>|(W?J!9gEr@fs|8ScW3ZA#()z4RjR9do!<%#N_+
zO*=U9(}a@)AIkKB#QdbnliUiaCsA4?8xEiU<{W6=<PG{;ShX++XP!sKV2-#pvbmX^
zZ~UCAS|*uCs3e{Rx>z99i`NG`;%`|Xj>&mn4l`@?Bk&c^R~a0*cWjhycuL{d@l342
zZS$u;En~eba*C?UF717(h5f(5h{r9Bty4uaKgRZ3-g-qmaH8ugL%w0KRK>!{d*;w1
zV~62H1|t(b;1-${Q|_$K>?BZK(uVzE>Rzvl_J|Y%92L_FcV^=|+<(-;^y2Gt6B~wI
z0WJo6x4#la1BKl=cK{R}#*c~N8$b)R?i|$25<Wsdb0z%Mo6}jFm>)vWF!*LehECvd
z)&VEqS=7<CsQTr)#ml)Q#OVOkqy%mgc!|S!f=!lF9Vt?Rl#(4jVV&Js7;0G94T6<Y
zrfhZhUf@oh%*5cYfUzddYi;j!)x-4GN3aiKxk#U`bFVFko%+v-0QFtTfOPpnS=EPO
z<aSEiJLVai`AAix0H>n?w#SdYq=_w_zdSlM<u^f-Kt(lfklfH2D8G2?{)8qGm7RPb
z?<7RnQHm(e#)LGCIJlj-oRr^$e$)@4D1&psYkB_j%yg%HaSIg{^JLGe6m968UHtAv
zk23;cOntt55bb8_jr_w_FK}btB5vl&qrckhOZn(pf_KEzFVvhuk5e0!>e<5LGo?Jj
zB6}1cYJBk{|NEA#^#Z}@x6#xR>seyLcbEURn+YgES3Z5s!1&*AXbjBjk>e%TxRklZ
zR`V2DwMtuox(zi^S7>K#m7M)DHK`fr=!%I$+_tz*P2431Q55-5Xt~A|E1sph!XRIW
zLir%DoqT1(cp$CgN7*po4OKeDSete?gp~)MmocuIzs7wIh;gcwJH+5JgLqes$srw+
zecS~Je(IO~b@e1t^9_fj0?&v3l0vs;e0ab<_0ek)8}Xt!837Iy0@2C957GPWX?Qsp
z&izZu;z7i*`Ygkz7k~6pK8s=|me&~wDYZ^C$*E-TvXV1ewF(Lq;6C2wzKxAOOst1k
zwcgRy4OhTW7UiC1)!vLGE=UDilBTiesp7FEV&R_slfG;=)2}As3x?zt)1~ol;*mV$
za3xXyjo-!N$2!$<T|_R&9Rm)Jeo3lyD)CdF%NNG+h~)8`i%Z3a{9V#~ps)GjQjcV#
zXDk{^!i4TXwN@ZFQJgKAKFqfs?R%f^*THcaY`?y^IP?^{O!VP*DWM!?e!MERT*d&{
zBpLNidlq=WvaYwHBQnj<>5CtHVIzM1<AkbR%Nu)Uw<%+e73XWI**A)!Ke1_Yy%Zjz
zdMx)^Hpd|_g@zpQ(aBUX=QC+Ar#-Jm;jC13a3pKipeSX@Y2tlFA*o_C^wNGg*=w#&
zdoa^UVto+lrwkRNWeT_wxc#)O*YTp9dEx4hjB;inDIs~GC|JY=<!rjEGy6$xos^%p
zfQ^i)4VABOg%zGDr0<=_$zyc<@~pb{^A(^UrgDzt+@s`$xRGoY@2WQ>BFUHcm^y5o
z>pku*-rOx?EepF{-obqT)Zlst4;ww%rFsycnRDT)(eHYaMIo@l%NdX`40ISSNyrl!
z9WPprW^=!J2<z3fsGpmCT<ho@BYG{Dfzf$o4Z=){tjxf3;o~L!i?BXws%qVYPu_V0
z`3xsFNqG`8^U@aclt>$FS_M@%-26~sdfHDq5W@^lsU;)W-qpgwUKhp$tEEL+$f<uw
zHruXo`H@?Yxg%Zg^Sz)PNzPvfef|&E(3urz=E7uary3d3g;U#=p|v<%l|l!GOW5uh
zJJ7hY7Psc3m?HzTelwrg5Mz#JL3s1gi`}PfMcv+K<;W%JPtTv98rT$HIa7e+f$JJG
zIN_J&I(4*GZ{Jg(e+X*k2$!9iaRCnIGHxxTg4(zG=<c~zo<fb@jKx)i^1YxQBQ-@H
zS<XV=Qwk~C$kTS|#)>tv%DZ3m;zjqW`yu=dsK)?l<-sPGNv+I&x7Ur+VSdL$wU*a*
zG1<?jh~gwlcRO6b({ta0FWAY8<L&0-*Hm%|TKT>n$7;FPYZVpj83=T=w8aeYNdnN2
zHG-~WtM>-gTB@f!W?T<8hD8b{bMCwQW66`yA%3Y9?-ohExevDuqkMQfkF}|G22v8l
zOZLWI)~2+17~mf!|H?ZEi4&O7pB?5tE_V7wcy=lap+I~|s;+0>KrS>W<>0a?`jsyR
z!uPc&xwaI;#gCnijja-+hS)NCzs#DnZd$V=+aVV4B39#_)rOVNR1Kv<6nX4~7UuhG
zLd0FU=L)|<RH$IL*fnAwZ#@YUnD~@r;>qTBPh^=IR&eV)k0SH#BOc_s^d<Z$)r3pz
z9o>^Lk8Ck<u(V~dYW`m>Z2mD6IW>!19p`hp3Ba7`Ot*gf^XMGyP~JK>5zeuoC#Wyq
zVBx77;LHDsVuJqBrPT2Ve)+n$jp=LiE&JU9?Jv@A$6BhJ>+OGJ@9%-L9|N5Ou=`mO
zKzooCB|=3V@e8$PkdhzS+MULLDc2%j?0L$=xAJB^asvjuNJ;i2!t(F4;|b<S%|!ap
zU;?AmiedGDK05O|?{(4QwG%TV#7xyV(KeGFo+*km5LDfG&aC8%+p#oSTuIdN(hN$d
znf`Xt^Zi2MyR|n_RSBNl4(&6`*kA=V1v9ahg3m2qgtr<K9FpCp%*^zZ#91E==tRi(
zpY}HUG+eZF$jdDMZ;_>wgjko;Zp^y17cCI|a)=7u{Y#Q=4K~k-NiBE6KHe;A=CJ_G
z4yG@A$7Q;=&6oMndUtfs8Pm^Qs%%zz@M-Sdq^S3oM3=|I`rkfP@QSE^GxeW7`zM2g
zn6cR#OKYZNZ;63tjy{P#pu60oaX-Pz9DHsag$<~Af>tJ%1D3*3`m=n?C5Y0{%-9kS
z3wKU!ID(<u8I-#QBJFzx>gIWYFjU`2Qj5PkS7>#$4`%xqNl$_%nV3dawRui>4R(6B
zI3oI~&HOVt$#{~57B`uok9Fvc(PJu;RI>4ZHW>#aK2O6r;<?|CZvZS#zNv>?{LvsE
zgk4}lAB-(^TlLd;U5gFFp39N(w#6A9K+mHaH(oy-0CCJP^)`522W<`$?guQ^lx{Y3
zi^&RxSwjF|%`;sBUw(j4qytjw`FhF#qIfC!rn2U@FMMJY(qk?l*1G`OV*;=R!qVtC
z&q4E>RLQEdH`gkz8oJk6j^)fsGQjDf(Z!;r#1$!k)}*P?VMV|W<!VMsY!no^`&3^E
z%ao_^a!)n_Qfxi+fncG22pI`XEZa&yadRI54gTc61!09vW=NhXd-{0>ywJb-h5E~|
zoMG=ERx5bf$U3Gj?$Uloj7NJP*Y*Nht!c!~Q7LiuX~DOjm?{<`a=PrdxSklzo72;K
zECmzxj_s@<U7;o9PIt(eyW^4~{<ChQ)_q<=fB+!mni@n-t9}a=YLCM9*$w@GzCpyv
z01jVL^}pxOUnT|~9sx-xF#ud6vz6ObZs02;Lkzt4;Shhni-6(6zx6FB(U8$WB$unP
z1@V92bgG1ZQ3p$hDr+v(-qs0<jsZr$HAKb4zbCcCXtpXp2W=&ue+av`29%rT82lha
z1@;a+Oa~oVAu~5NmnXvtB8_D>tNq?dNRuJ78_bHI3`Ns&gjsjn4uh8gR1?@+irVeM
z0qQi{We?c!d@nfOBZ37Aerd!rrgMqCIC{6<_9$pL`f!{%>1vde96R$yn7aiO;_(Qe
zMX<P4&NnIlK)z@1SD!C1{e5C%RrYV0vGID`^aN(pukHVLpH7sl=$w6=$7?9s3m{%K
zB6brJ(%68vnOmvR_HmVhsy{R{cBodz#5vLY8f-mZiySJLq+&8{g356o9&0?ecpyB2
zJjhjP!-kbkgs6i=OwJ)Xc-#PQBql{@|E1(R>ujM*Y?aFMsc^x-1ZVevn<vD8bOffc
z-h^dUZ4n_SP)+|nLy_U#bGY8>d%x6P*UskRah4=&8m>2Xi1-Lxqt(kfL<%&z-2fYn
z*H_+$cof-YqFGvE(5B;Dk~E6!JZn1i35TonT(|pESSgK6%-8M|X<{Co_Q&Z`!)DfR
zlQCznobLjNSne{yagkUf7FqYhi1kD*=f#hWe?Z@Uv!2&LlC#6uK*q4jz4M-E)BcQM
z^WAjRn8wD&`T^{pCMq^7m=YwkWvP4+53@(~MQOTPNU5f;jKu{E{{~`Xv*)9rG$=gd
zfHYP#B^mmZLtQ;3(Vk81E&gj$bY7|z2x6!j?%*D)q&*Ed44XP4$8oisdE3%8z(xDi
z#RS^l@-D#76aaJQj4bG#L*2AXnLf&!r@3Ul;HgijrB+@?u+;JNaB<Pc#<9F|H@0X~
zc_)d`f7I)q`V3il@=3>R@E=>R+*Aa@v~MN?q(l35U3(&dnQnY-ZlV;F_hR?qZrDiL
zloV;NXXRugF;ID|RTd(#U5{($#AwDGOtlB^%mG6L`-eq5hu%Nj@j#(Rh6r7PQ{3L|
zm87o)IXk3C@?w>E*bAa9rNA!AH?np*BxaD`_}?7QQ_lIEz1L$8ED)a|!OOmwxrdmY
z)gqv4#n$L27YJ-J(YQ_MNw`d_>$jOSHmZf+l`1SA6I(60jbq8*T@jw038=}MljH&%
zyc19K6$(b)oQPMZRDg2(3s??7X2<C-Igi|Zlj=`}9A0xpVi(<yM<AFnu`hoy&1MJ8
zAj0kS$ag?>pmy5dTQ!o_W@0A(7Q?3)>t|&!<VA^1>{eYnFPnw(e1A&x1r58IML@Rv
z+j6Pw8nKiC{fj-5dOLCesfYh=F^<>c{G1`G?3}=Fq^}ePaxaC~5qxOpK$vRQ#g1Rh
zKg;7IF)tsoMM>ik_aohF8w3oDPVp8-Ord6RB&{?s#aW+oEk<ulSNzxLZgjD8=J?RF
zEOTDx=%)G*qxN?$s&4!Mk*kh)zdu%tPWTx2qp2hc&V8rART!@9$$HhVf$i6Q{EfQy
zUhw#>cpL6$&3H`%44O{W`(>LR$xp5f<}9L(qDcFQ|7~a3<L>|JJV3I{BbIdFUcx{o
zI|p#F2I8pWXmXS8^(L4H$HU~V1rlJPRY~o{BODO3H;%twM=EL-c(WgO3A0%&k-n+N
zj(dp=JYaiqj%8tFog)A8)I#$2ZGH=&*m2nC5YIolIa(Q+DY(e$t&~x|aai83V5W@!
z+O8WD44b4lV?s#K_#q9l2Q*HQN`ag7R-6`drpUkQGa=<V(|R96K%My=TOHb3R^dTm
zGtLY0z)F*E3f6ouj?`i<6H?iP2`l>CZ~Q=?|BE`yaLh&e<x9>J-26?j{OQs<)fExH
z(z{`Df*?@?aHE2Xy59@(Q6{^cxfAUY;+XXer{m_3d#gn*DHh}$J3p@}{iimD514W|
zT>hQkI=z>&l0e%tgUssFP+iqePMuy|Qryndqz~q$=n21(O@@7XkV;T$;hoz3MPe38
zkbjJ_W#B6k%3X#ihO&<%3c!Sy-zrr*nKKii&zU-so<GXt1cF}%*2I!4S`e4VyL+Vf
ztZevFU|8xEAdF$RH>I~EjV^CES^XEr+?K4<ChlAtrkwj2N6S<d4H}kW)>kazs!E$B
zHOf@0h>R;aAUAl^-P;wuI4<E+AGQnCNA2WfqJSE~TPJVF%$5THJDxA8m@nVEY~%!*
z--ca;*e3c@%P+j)G#h}w&P>i#m&6k{D?b}io!+|Mk{|}=MDQ2Z>1p0o(LG!DM03g^
zWH{P`YYYPpXAETHP%v1h85{pD25LOQB54YcXeC^^A#JFGQfGQ!|BJo21;j`7FvWG`
zAb}}}RI<^}z?XOc8YQlYu=qoG`n8QyMjuVYJja=vNNCPgM+5pT`=wRJ#w@QH7p(K@
z-bFIbh`|3~&Jb^U&Epc3+kezVmbI3%sd@70E<ecb7`&pjCB|maNAeU#joVIf`y^}*
zA>jli7AR$_rMTx^71_rEP5#=kVVZ{n0FIo!_4Ialm$_=kUFR3RGk5$c_z%Kv9|`3Q
z#8UJq0lQcRoMe{B^|30L(B+})jwq@)V3~F!sX8hnG0Tm@ZZi-%kh@v(_JeVv8-#)R
zW{m9vXg-n<5g)ADEqU`j_&%ljI0X~uv`!OSOswJQYVd}j5uYGiP(Cum9{}h90nQW8
zDIY8!iKSAJc5jj^L<?k>k7gQ~oCn3ba{tumG9t%(oycd_;i=wLnd+nbCH4GTNRdT}
zr<0Kl3eL(2As^k6J&KQ8l=KU~@mxIxoMn9l={nfR^w<2mX{k2&CA;DuZS84&epyLC
zV}>T5i^dWGo?Y!+W_IV^{-?O(-pSe1)mD4@OAbgn6W87NYWXCWcXdevKTk%ACEKMo
zb9}3BF^M>ZZ?eB;cYxB?D%BE$h9f1(hb8wm2{lCNsXU+cZ8rH<2O7h<CirC7k5iWl
zJ#oKeyR&tN?rkzzm7%H|`s6Y-n#zr7-pu4ad_XxbWu8s24`2H6jPp)0zwyjaseqty
zgB?H6--dXM43GbOftM%l^AUO-#WGH`7V9eRmEhpiN)7jCu0U3X#jxrd?zFpqpkX=N
z3V!u=gQI`C84PiF`Q^pF1W;#N9!DwN4E*&ScFl&)-|2Ey?pKy1sfkaP4@X6VAV8Xb
z3l%~(=RV54YEF%~u+*^rNP&~@6-F)uy)1n)iS+y`#p&$flA-TIZ(t=x9TGZeSDiO)
zH1M393U&>j$a^sA^hc6E9t;l+>W4O^)Y=`9em!f}ReBp-=J1{p5nt<urnrdw)72h!
zoA2IFLC82~@6Ds+*BVfVTPX~GfdBE{S$*b)6D;}kJMF(RLY~n*l?_QkeaAt*A$=&Q
zR@c2iW(ZM~^t1@p8?Bh*AP$b)<Vo0*2hnJBp@%C79?~Zf0tH!1F$q`Mo;ulyx_b67
zn+3plwV{2J8L=XtugeZj;7(82Ga;Q#V}JnmNxU~a!@0`{kdFl5Ot0rMnm&7a?A0Gw
z11*+~6;bOtGnA|MF`AeU>W^l5{eRwLI07P04>RSlc4uy~9oIK_=GQ~(Z|ZMgCN2aG
znc|afcBLP2rKe3S5Z1rbpZ%L6UmdeALCetpXDT{8NE8WiFh<vGp4~%_WZPBjxdp;U
z=E`&@cL+z7p82yZ#p$iTdhDv+!WGFv{T7wgwMUh7LXlJMrDvJeN+<x<Zwbp~Xe?i%
zHJLeM80Y2+u5>>#Cx?8~3d^iKJzuU1WCl)h5hb#Y-fR5*gt=6rXb|Rbw1}(&r%tM*
zhlL!tq+fr3!tIaGbH%5va(1h$Bh(xRJ_W}-ne$7m|Kxtl?Cvc4d`jAykAnfwrkeY+
z;S*HsElPd1fF|(~?OV03+*B&1zLX>{bwxa1httRBHeO2}kVnC9W>cJV)k$OP%o09;
z%B#vk%|DFR+PG!&1bMO-+FksE9}O7}{L%$dMW3xf;GGO2-dY0QTHdIp{9{`$x*kz!
zpgl%=HfVwY{31SCAyDksAH5T_RN;Mf1!}rY-u#}Rb#4rb%}KSlGM8#TUS)Mr&BMAh
zsG3cDK$R@|l|eJ0ZM^zv_Rk)DN-@eva_2dw;YD3T9FrH>za4F2#Rkm%O-DNyQ*?-m
z!sJ#ddtWci4;KQJ!S!{ZBi*4)NBH~zSswF4!)-OjrX!^<=le;7C7Sk2=b{RN*b&e2
z;CuAv;XFOCrg|hfTwa4xQg<c}$g*XLKh^wUss1KR8yy|xRt0gD%x`aHEP^Fh7e?~W
zRlE7uykUFB>X1+$!9?xJm8SQ>EAy(qeVVy+V3u3#>xvF3TGVNAxjB-Ja~Q|auRZYN
zWVQ7t-|0|oFrt|ds&w<n>Qo@7Vg<^>US)`jj($YyvFLOx{kT|r_75*gm%n|~dsJz7
zr-rgaE7$q&`1+?V)V7ZpEx4V5i--w@6+bCX;aL*=Wj^Cvr?o#e@L_oqBd@L?XvcZC
z_Hc2yZJu<r0yX~EL#}P4hO0mkQ+A)&-?F6_brbLq*bPb#N1DCbUSx{0D`=z!Cnkuo
znY_#NaMZq+NymxcKs|Z9Rs2IDS@UJ+ZZKaZ>x`$-0iTYNQsMg@-*#Xr`8Z)BM22zi
zyFaNoRt%-OeO$n4<R&4cC6xLDPO!E9ZlG>Qrs8Q4j*sH@w<NRlZ!)mV5#c#~lGq2E
zU6AqRzif45yyWClgB8t*8MFi4{zj*d$27!fKGW!r0jd%EtJ(W<YEM2i>Mh{x_OmpB
z80uXA-^+|JwQ0OD54)KXJzzp*CnQ6@L=T8#E&_jOn4Y!FT)!bTYrS>-9JlM#@n%(#
z)7VJTKHp!nJ5yefTN@8I@|L?pf#*|M3^uu??9e;=t#D_CZBMae3I3@|`+EE5v&Rw5
z6!M!2zCYd;)W+6-CW}-XxjG^{AvW4T_(stPH5%4dwTQsY7KX9pj9kQ<NNqgcR4eUt
z%q>i^k5EZB(CxRp<j2d8`mE8(;(k!4$iNY*;I#;6<n7;o)pKKBO(KIup1dFh^Um4(
z?6$k#F+1?#SFBmDVu26@i|Wst)P=ZnxU57Ze5H{M8Lb^TG)m?RX2bqDYCMtaD-J>r
zj8DM<P8DC$qFM}OH#y>ncWCEB%C{|Iw>TH3Lk&KMW9RnXPfVUuUS7yc!nz&p1F##P
z-8}l3bK0^ybK3TQ2ug&HAnZU?<;_s`NSLlVQFt14=CY_^B$!=0eo$bNB$2cD8M^4v
zF2QT4DT+0iBT8{wbkMd!8xR^lf0RAh0rVrk+X&#v?pG^%P_<Q}-l$BRV15l6M0}TB
z=1ivK%1xz=`<GN}k(iQ5B&~fbpKFt!z*YRydRqL+(hpm_9O0^xf=28s-~zJj$jetO
zv=c3POF5}_1C)xYbo*C_g8J@neZQ2T&2QKWZP&d_q3GvW9*8Ul6hs9CVBS@_CXL=g
zmOG2c%Hh^q4O=rZ%jbIlcpw-9=<YUTH*{dH(Pldx@YDvi(fUC47Jpd>tn>hDS7s}K
zSWB!CkF6*+#JEyjp8YNT#kxy6o^sHCT@pQOH%u)yceN|zA+m$3)p`$K2g%v6jWOGh
zbtnN}Ab+PBytuoGM*)pY2~ahdi#^e`@HK3`6Z@irDtX)B(^Teo;8mG0=TG``Ml&@y
z?IkB!`k<vmh!dL*?ACzaGzw5b|7bWQj}{48nn^o!>)mJu<bOPF_7s11l2rXYhoxpf
z94!WF;ft-LBd8LW>C{Rw=tqM<{DcH_b99?c=Xl5oUzi0&B5VW+`894aMw?BE&3)EZ
zJ|p=DRKbj#t}XJ=*S}M*54Te^+FrP<UF~VaEFF}y{QYsKGW+6lstZpwfoSBVmnPw9
zOxhgtub72LSCL=`TBQ-{&$erC7&H-X0>}0@QdJw6YoFn=o<(<W>YZ>#DpLsCPtIKJ
z67JFY4_F#1yJlCG&hzWW-`BTp2wg;KRrg#ty;yStF-`p-akrh6NQ^wzz#~;f-joz5
z{NS?Gaou#GLjb;nRiBE1a@lanxLVEXnCjq1=t!~UWK2>WSRy=9Mb}L1mq}kHDh7GU
z)8Qp@HhjjfTd!wxqP5(j!WAm?ng;^q5QM%DtzcWtC>15G6zAd8ctMfc@5~P@zrp3n
zXhmhy&x?JY=Uo6nScm(JAJ*eh&Y7dpAHSw(k>WGYsWWH`4Ec~<f8;;FXg(x2+$Q<z
z^-s-p&tD`d8sCy#h~FM-{fI)x_M(W-D7_6|*blkfFdUqGxvFK57|#^rOI9q<Sszf^
zZZaje*$^FmEcYWtN{NAB%dK4D&cJYU{?C$Aogx#PBRLoG;^ZYH71zQ&O!6;Kt(*3*
zQu`P-G+d<L=&Me6+5U}smiq-miwDxuBv88Y!_}+ppye?VRg%*3bQs8P$ZJFSY2C#`
z;!(N1FXid2QfNKr>h{U6*Z)@+Wt0l{&<sZlTGQ07OU}(i=hPgUSCyBR`#9=q5d5tK
zth}d5#eOUE{g)k1;5Nakzu157l`IHNP})}5KxQ=9j?MAW1b+`3Wxl3Ar7<bXM9K)`
z9+Nm9s&#Q44SV-W?=M4MVKwN<^T!RZRH5s+2GCGCNhH}@Cv+X!k8-^Ck~?iaRyBd7
z4`@;CcV1ZjwMZMm;S7~ZJ|X6bpKQDrIC!U5nA(I)BU1s@6VGVV+kb|K#HZ9hT(~Ow
zPIDA|&rM#mH?L<qGP+`fB$MrrW#2-56yH)$oAel?oyEe59xKuW3<kdDVb16|lN)Oi
z=@b7-&x!y1U?_~J=U19bpRxDtFaF8n%>R?eoz_&%O45*8V|H@n6Fx+$cx108)hMke
z9iU4PncJu!cm?uqDpf>z<H`-6S4|^kDCt@O9p@xu!rRqJp#L$qV~YCXe)_BRUDrE2
z8msal!$m<4@rt9=A(EP~G7&G^9O8%BtZjRvVz9Qb1`$fTjuZZzi6;SF%jLSgnQam&
z&6G)jBs-N#p4(Dr%g9S;gIUk|cfUK=5Z|iusG!5I3n4e;HPO8h{}qF8^r{yaDH>{7
za67Xjapte?5D&@i*&8T66$|2=tr=w9QSlVs!N~WHI^KqueF=z35luhX<%OpBlJv4w
zv2l38?^#&j%2Qj;z5<HRKDQ+gSN`70pV!uRHpO7dJuk`KEgqI8H4nwg{*wYuK2@3C
znd-;VaKb`h?+6+4bxW*%cxCGi@O?F*dqr$ituUw?pHK+z*Ene0LEqjCe@e%F7f_4J
zyc;fh#n-J)H}#u@$Ef!^B01`(V)?}84C9@|R+fonXRd0;Zv{^7RN&qZ)-v5R-K6I_
zH)|Q0I=s2JP4*SP4U>lx`29Q2{<YCe%Pk^LqGP#T{+|@GmchjkxTux)b=CvJ5zJ0x
z)MlesjA}=jbxz)k2ChPuTU(B?rsXd?E|ZaRfjy4an-abxkpy$oY*nM&1xdRuH>n~y
zgft=0y}&ar)?{L^x>=-4YvaJIWO34b4*B#pLt@6SLf(`#KBd?-SrRa7K}BMy6gm#t
zAaGOWi#)exiqT``8M2+3{%5JDY}RYM$4;_%X-m)QC$1+8IUOI7Y!}DTJq?AdI&Pqg
z<g8rz;)+OKj>-!68_gxF5Du=PJ*IP@4NdTu?{w)e33lgm{o<ta;pb+k07fI_B$VC{
zFX-TUvF`*NE?~RW{%>5k9fku*3w+Pq$NohsTaWfsY~|;KithkfBz%fDQ2_k~5m4#b
z&(h6y%QfEbal|z+5*@*KLW(o~49qr1G;B!xGrD5<P5|i1Q*ru-*ORywoHs_(zw~Zt
z6_MFHmskyVp)z_Nq3(<Y^5)z4-Ai-vp_*IUjIdC_N8=GjJP)Y6H7DLZQ*Tw_kNM>4
zmB$=yO-H74YsN|&C}Dc5oZ7nY^GV*bBf49fs;Qgz!VqzH&vFOeXN2>(L(3r=JH7Br
zZyXT~gr;^xMjg(PCFwPG57x6+Gm-yX+O0b4w$B%2HF9uE(-%f2jA%O>DuZcYX{BHs
z&b?TMX}HGMcDx#l!tefjOj%}1G&{!eJC;PUKF~Ri`Mqb4*n00^PJ2y|rDqtbf#oqj
zFBanbTCx8qOAIV!a%|JuKXou59qbB-TtTnuuAIqb7v|X<+%KcDDt>IDH7I9g&1(aO
zT{rKZTL?_dXcJiU7CB_I#ie9UM(Z9Z$e4Aok^nAV&FC53^UQ|Y`h?r-u!N(PBWx8_
zt{;6hXX!MvC*h+pa*LqfAN|Nl|7i8eLW~{}swT}QOIOK7RA!J}iVJc+t7E&{5HJ1%
z<NUIYI5$=%uj1eAn)x(bB{Tq^E0WLsG^?6~y=m|)TX56X6xE&xgLhFrmDq2b`<q(B
zc?X`c6ZcQ-nRwi=U=3vhte*Y{7mZJ<d@mOy)cX`g1MU&BCxWS>)Msh60~Q|T(&{kX
zVYeTVFRSkZ9o)R6HG3-5tOF4gu)D<&Fz6U^x@63w7QM;WO^JB0EnG^yn{5Qm{e15;
z<BU|d<7=ioo@F=9Zl0aF8J>T_Y|+hrPY;GjLbG89PxNyv$gXni0_HrSM;dv!+FyRQ
z=8U(vPVM|}qYlY6I#7GEC;qF-JfMGyW;c8CQ}_Q{%EGwax6S=sGle(sJw@y+&U5!l
z_b<y7=XVR}Z3lkWXskCo7c}$KeZkHea@=x~XvJAFU6>3E$}Jl6Se^>^ZOn@j))tq`
zH<l{rI1DGmz`f9ArKWzw__CP(!q+dC&CEZz>*j3n`^VRR3TPrLboPa=4jx?^7*1Wa
z8dycRn=o|;#G|YsNex9`_$^FqbjAl5(PWQ~(Gjw16nSH!YO_Hi`H5$(%a<^DMmrPg
ze9WbJ;rb`j$Gka>!J}bg&6L4fCwIIyTSt}tB@F7#fY@SC1aOTjo}Sv>@tu8j>B{SJ
zMX2fLK%)zm(Ly&j*WIGjpZ#(1{o3@2sr9)FGU4>XE3qEvIl+0TUL}m2o=u`rZMYZP
zlUMrL<m$8inNbDqFB0b2ALWQf6vjrZ^kAj30LrksB-wZ?&=W}y$cGe#DHutp$Iv;u
zq$J-?8K1}G`sc-tN&V0SpBSaTc~|Xb2opM4jp$@2w-}7VrD>U5@XM%fZ1+-=I>mCi
zp`i*VDR6rxIVbEKG3V0SWYR2x6ybAhuL&lJrC6COoZL6>I9y25TbyVEhOwx}x(|FF
z`@c@!+BAy|<&>$&MDg^Q;l;@6!g7l^W?GH0swA0NVd^*2P*NtM@$ki`PE&R2ZGp6e
zkIhPrU;=D+C?SUprw_SHKciE19hNGJ>x92miZ|Te-tTVqbZ46)G{?S|+uWGD<By|+
z5!j!^$RSMs6Q9?f;=6zw$a}U4CDJJ`w)$9);_|g}xoNE-#n*qNOfTXVNA@<CX@S`W
zQ(F)s&qc2}NKRYr3)SgielT}e?72`=33wi$R?xD^ylW}?>}m-ly^C`Fu1SSN_;_5y
zk2Bm~(0ojo@<9&>m)&okaG&{K3RgD{(FP1B!0EB7{N4azyf?v3AhU$$Ra{}8YRH}K
zpvd07G__IB3M%Z%QxN2Df5a7j8$gW~r#M*-@&BB=?##^XE<Gk{c@$Z14;Q{KzioaW
zokWJXCCRQD%OMJCD2k)N0|N5tXwmk3)Iuj)$>GIte3bBoxXuHO+0M2B9rgSK<qn9N
zENzl8LLRs7uK#t1Kz)a}7=+mjNG%^1mgEA?+X_cMy*%w%-G_5`jnZfI9{hN;B4rZ{
zGJ}NiIgO(b6u}hXy1yrL4sPaQcocqyD4M0_m4B})yg13MTT;`o0#08Gv0K4T4tgoD
zFH|lViT_Du%@kDsm6VOYB<KzR5pu@xfc%DxotR^<@baCv|EH<<j%WLQ!@gV8-m0j*
zSMA!hiB0XjDk!mM?OC+7w_2^eH??Q&tre@N8N?=bcs~99?)!P&{z4*>T*;O5I?v-c
z-ax{=JA)-QyG-0IO0W^Lhc=zfD!au1Xzsy_&S+~Rx8QsyJ7#t_+s>VLspMkQ9BlWr
z@qk>G3-xbC3>9z)+j_pRpW?9)PAIkcvC$(&NoNQ3J%)JaZ+xHDg+OkA$-Ltm1}}ND
z&2DE(h6y4ZG=5oZAq=<_R?sqH+^Z!C7E9q@<30n35>nQ4HIr1KDu}+UqU^54O2E@-
z?97=TF<il%hT0z~5{39lBQkyrOCrk`-Q_WeK+#0+y%mbPY`<KShRSt?-`T>hG3mvO
z`h^&BNpH=@zF7hoiwbr2M~4a_as5wiUG(pPnt7+Xc_)Cd-j>FO1Sd0zi^5>T0r!;+
zFUugvL=XGM-r5cq!g5p!oRVK;$bSk_b3m8>n#N1P#^k7LarDje#YlS;*2D>+Q207c
zHo+M<FiOdhTf6GSzp#p9pZKjVdh1J$z>35{ZD8JcrGJ@Xm29tLK6*Eof}vJAb1Wt@
z14wQ)DL_>V-NccfmZyM-lb?-AxpEuR5$c)6w!v!oOWxMI!5aFY5bilXJdtPMB4+{u
zcLG1>{Hz5eP1i4qZQL{KkmfzAf~1w%k?aeIm?=rg@sFdKnzCthrx^g>FsQYRP+sFt
z{{_RYr!N3wa{dW-91)QLbt<wt4vsHu-+g!0MG`OQGP)>-dJieP`!$)C%lCT&;elrP
zq5iJrN-7o(ma9G{I=xtPc5x~n9iQ~E6s)^Y-R9|_rs8TmJJkl^(Lb9XkZH#-89+A*
z7eUa-dYhpSyOdxCC{SNv8)4(L9!5Y1TfGK_`sseLBS&``UKt(3BoiTltACP`0$the
zifE$~n!oxe@#f2r#gHyf2}hI~@+?@eGoD-L(47_L1=KNp;-{AoNK3-kkqNFhzW0d!
z@>k=9F7#8-sc+6b&Q>wXZGs?xWDCe;bSih0Xjh;?A#$32;WR3^^io3x1AdGWc&Q=G
zKds46PZ`SKyUB)P?q?btRjId8ex^9zxVAvf=;T3Q&Du|y^9C(0r`9W@2SJd=h=z&D
z_^ls#8*+Lie#t>jVMDX69H(d$7S5Xxv4eObQN-!-ahKjmY5C7NXCw8~{lVz7k=MVi
zRRO|~5HVfu?Vk}=R(Sjd|5u&8BYdZwSCGXxlH&I%LAqT43ZY)CF|h(bl~}7e_^am=
z3S|~oY62q-;T67@sL6gR(ERG#avTj=I%$2X=5s1;)R^htYZ|>@zkU5a%~DwI5?AUv
zDcFQO;%h1{m;1jTQ!*C&vOa-cT@ON-d%GTbS;Ni5LiL)qbtFR8Q0iWKS4`xrhKzL=
zV{5%WzhL%ev>{zXy(yI(K{miN4BYWtsfhdj>g`&KmQ@_Ker$vLxasAxXqG~`2D!kY
zz+gQ-y$4nPtDGVZ4#uTY`3S1t{`z!q4mwGmz(r?ZrNv}km~(}ae{lL7gB~T)W-etq
zgzm8fR!WckqGlE?n;X?r6ywE4H5+8lN>1=p{2k4^0vDkd8EZM?XgdUtEOZ{!ND|C}
zhq<Kp7#rpdor4{>D50YsKn$!p3-1t0+sBrR&D2U6X2*oCZU4XKjFf}=i=n2E{NY0S
zNL{Q9>==l0iW<1C6CO*=Bv4<`tNJYSm%6p(t6AYI0sX;Oo5HuqE)f_ImaGRYC8jA`
zCQKpPWNj&e7nc_Z5aw%Yv98+$W@A3ndd|gZME(X=jLtS~3wGKPX)>V&s8Xd7>Vd14
z0<&IP-p{tTuT$FVyPOKAsYkFGLcB}nl$klayUe8+9AC(AUi~9WeaE8QwF4_-ysHJ1
zTGxuavB8Lt0zC^YVa$KKRzjDAr^RY=Ma1^OB_Smsa7A%$LUejj<6#e-af1ROBC=y@
z5d)_YyyK5rYVPP(`lp9DGrjr^GZjb4Of(Cl<7YOxqK6C(OJ<r`6K~p9g3uocMIwS?
z&7%sQhb7$7+Lyvmka8FDyON4m2~;vR)tkBsSCV6e@uNc2OX1*HmA}*n>MEdu^U$PZ
zH_KlOYM0Up?<cYyU(8O*Y#$fZ=YSJ>21V+vkzLdjSJ*%21(HaAtuAl{Dh&I9uom1Y
zT89Zvr_nW1X~ffD_IH2s?*m<x?rbQYT7-+tu%{9>OWJ6h^FFof;o0*a&Bjy@!)g9=
zoJJ@V)M7ho1BYjCVG#LcvY>Uwe}ox`&yBnUq+L0Ut%^0o|0FR&t-3VdW6M-R1R?Y`
z)VH}Fxyj>SkHI@G@ojmqWc#Zd6aj1L;z9ZL*hC@v&x6hmz7t&l2@Wsih2N*1!ldGq
z!<R@se-anvJ@-#KV9ONr6b|{gdN&62WjLQVNhb#t(>wUj;xN8dM2;kG>HFvz(904Q
zt+!+R(JvL07aJ{>=|8jhZ2hq19Zv0oLKGv^O%x+;-vS7@NG2wDkKvFdh3cb;fnEbc
zwU!F+iwnA;61t5mKb+*OBj2<$R1B7-0&Du;sz8pIZiDwz)S<VC%{Z@3jHASuAtB&6
zQ2x&X3LNSt;RRh9HliH=m=^8b!(zd?9M*5Kvg{}tOoNrERJv1S38($kSQLMAjfrA6
zm*WoWdPn_gc=bJ9UF4}pO?75gm_NOis!cI9zr*m{d#tEa1LN-VNpWF{_AU)Wsn{&Z
zax}V(+jnN9tRGFwGKNPN$C(O-y#bt+Zfl5t(s$T(oZ;nc+NC(lNwuU5Z;zx$_UO0V
zF@1(EPqNJMbF}9xg8-jon#cR*T<Q^SFmLgAG&dCZ$HI8!thB~(s#Dl0I;(l?)qpon
zc><P)sjW#KLJFBgUG;O!@Wf|lnaaweM$-0(8Ga<tmmRChVW0wSG7Hc)7ZzlI09)f*
zWlDzUXNRl6NVzP{UM@@$U<~D<2J#*u3<>){`l*#qk)vO=V1Vvt<YQ$MyIy4D7_dKE
zd!2K+ti-G!<{49um&BsdH#&%f=qg_tp|hku8=;cViP;~K$S1b~Rvj0sn}+cN;yuYC
zitHe9qIF0L0XvX!K_FigGr0Tbd4jyn#@D4BRMfRG4jq0tuNJb-1{*r%MLds{HNvNd
zE&e<%yf)Ky1}ph9uKWq+rxjl+16(RQg`wg1Tb`&93YodT{+d3c(6Qm9P+sKOvj60a
z5+x{xKNvzk=932-v^g>-{pXMnOI4WrUqu}*dIpLF<PY>!;QvrrF$&C%ONpl<uknj4
zuvTyRvCkDxbbSWU_bi_AN9-oF-q2EH`j|JiJCS@z;y1djnm>=J-cu3><1YNzt#b1u
za7v_Ckc1y|pao$Xb&W;n4YA?K@+`icB4HYJecGPi$bOPSRPA0zGZI)%z403^_8U9X
zvTznDs>KY;9_t(i(pXT55O45P%xbR0(D1)o)c79y|Ht$^4(dlS^_akb%CA?aFBprn
zNpE(8gB_Jr^&9#ZylEkSZuliF0j%tn(#&qllm37zcl;RPc1teu#|S0$M8OOON5WYa
z*Wcz>YAh8NU!;|L-Y>IdPB#RGb}@d)sSF)oNTFqhJbaCb39SPPtepv$pu7G3itfjl
ze^k%lx~@yd;C2E4(~C;tkfM$X!tpJKFSFToS6hG5VFE)KL9kboRYlLH^!Z;ap|jcO
zfie><bs@f?y}4i5P=Rr)%DM}uCR+9R{fsF2$g6LSxQ)?Y>h_65`Hji^eEN#J3+830
zh=*hrq?ajt249Lw_b8}mJKqlbk`quG%Nt7YC;MGtKe3@74At@Afq`T8tE=#*b}-XB
ze1>nMTXn^TR4^(P(ehwQ$^Pn99%a))PL<4m@B`Jn)z(HrUJXY$I{2atC%gh)w}9T~
z|H6426v42!B85RCpXbcpKMkY9tbl}kBTlBXk&6Qj;@=Q2CVvq$L23@^4Zlp3zz7Q=
z51;X(ZlZo+#3*p^GHmcggV|D(lMacvb%(6E;?uY$YYCC9g_ONTjV#8>WVx-}(-=EP
z&xC><X*1Yu>o$F;BURW~7k#qZ^bO<HXtN_jI}hdV&5MSsJF0*5)@7IAUlagx`Tl6%
zcPe;0^8FsbJFTZ4_FkOGI;I|MqD{X4&_|M6OMi=`x}o<)iVUYwvTllr#kY((&stvr
zZKnCBCMX3Se}5dUn*;r-0{FFs(`@sMZy*JN#OR`ZM)dYi{Ilq8N0Tnff86QZtAJoR
zq|fp~71X-TG5oJkwoe~RQa$J$Hz)W^rK+OB4r3PZvSL{^pgVrnwNLgTvyt(n^0(WA
zR=CQ)*!L?Q_zaME2+ekN7<RDH)4rck#z7kQEU{mlar?qslnQIGz#($OA|Ixz-0hxM
z9B^1MCU5#(84m#bWijj7s9Flv<ne3KDJ?$SCR}B2T<lDjdQ=F8O}jS_(bFjVzROKg
z&rlsAqDi7x%>7P9hHdP%A-j9e|3DfEC4uU3wmGp&e|M9uQ<LSN&KejhDn_<O0@d@=
zmy49B)<k`_G0hXl1%Ib+2X5l?gW!hb-l|OekDv&rM&7)D4hMjH5w#DoM`1Ry)^{Q{
zGt~A|76)_d5~I09`*EE*-05C2^5zC*iSoBa#wu_q!T>Rhlk5YLbf)lCN??)dswONr
z6=y!d$E*ZxG*^kj=$D+BO($1pN}!C2#lvSF3yFu*8p9RWeiH-Xm5X}8#fKe*Pz1L~
zc~EBXk5#dUUvYUSSQZ^KoH!d?l$eA8KT)4n4WX)%kM{>rwRUdCi3@pD)7ZhPddr7U
ztlzx>_kduiL0jb&)k|p9$cLv4_yrz8n?KA=;pTu&$Eo!nV0QOxw<AO^Z=a8!M3;Si
zev=M97<IKtXmr<2OGfyQ<1I=hZrwJ~%-!qD>PVNw&;sTJO&cYqn^8P>@>GATJcQXy
z{aAN~UrJPYm@v`so~V_QVuFZ}a(Ppoc$ABT6}2xzTUK73cHghT3^VHrkIM<in(2(e
z1k~0ophC-Tn<_4s-oyxWtcHyYT(d>Efetjks8Z7@k(b8@s_Y<uA9ORKW~|EojOUL1
zyoCvM^mKfWcUnCn0$;Pr>v!|aX&l}sQ)|y^>;2rNe=&SctH2~YZc^``sTFWZ*wlA|
z|LbNAJ~6sJQIxEwptSaU&uZ-U9(neagsCz8oyZ$zvP>#>K|7!_K(B|tz+n8s&N&+N
z4<OZfr+JfzcN06R#;;%RjQkKKdeo|vRH+chC}Bid@uyQBmWXlb=*jI`Q<g;CI-Oh=
z%NIuEH0xp`gp^Yg8B)*mqNYnAV*LyrEmfEYOC~5V|8ntPvoMsn_`c>JNzyqL7-Wxt
z{>bedJ7oi^z0SYSEIOUnT=Hh1cWaMiP~?ZH;^2C5umgTs(uwG57_T0&U3=nU&>t$V
z>>#Ma_-`s<bj$+2%=0|<@KO9<%6=9k7yyQOAMk9(-ihR$z3;4+HEC2{E_PI`N~@I_
z{dO?>{u7CjMknBE3(+kG&l3usX?r8WZiPG6;X>AWLz9MGCm$_*p`Dk#pXeMz<oTF%
zPzcllNd7eG7_~aZMSV9ubADd-pL~9%?;L-!#8!*&HRz*<cH3X3(DlT{gm;tT)x{wT
z&R_HuzlE5Kv2Q|Et+TOZGe^L&QPI6Hm4!yH(Onwip(2a56TpeT5-?_>4>awrsu;Vk
zDjCP7EC6r}@?db`%dWx!E=`yf3jgy#0^d0wkelZk_M)nWwI_G9Gk=1SkIs~d2SbB)
ze5~}9@M5LlPW`V?|4Tt{K=8!}UEYc9noGKQ{x5(e4tM1L?}x0Czj-*#>!u*F_5kV*
z^G-NrJP)r3KJs0UKh|=6c#1sG$J&0oFBtZA(;%K=3?K+}%C<ZU9A5uTKCDs4THq1V
z^FEv`u=vjc{n*aUpo3?ZTx4uj1%9{OC$7u`?B<WAI~N+Fkj6<4IysrjqL&9$7>*Kl
zJCTSkusggeZ$5k1tmP)nf0gS4CV-2580fyAK#+HOc`e!m2d@mSnM$05(i&qxYt|C1
zFQu|Db>$QIj5ylXk1<SwrjPBq=&O(-Mu}XVSoS0-vCxvIVadc2Nw3eo0Qp?!WJV@q
zk0X=wA^9XiFw<mk<k}!WnG|xu^u2g@^f0f85;*%KK~PUTkd{a&h)4MgwtLe3G<U?}
zhm#VKFMw<&j-rCI`g+>-fV=$wW&n(zO$@qkYNl3aItVmytx)l;sy#_8|KbMMGf7nN
z**zglW~Qi;Cy6;qs+5hoEQ>V>fVirccOe!9YCVY>2U4Lg=`<g%sD0ffnV{k>%}B3k
zd>nbi{1SsWM9s1ExN!4H;}p#J7S~GW_7L70hcoouV(sdI)<WXCm*uS>^wG(yACG4V
zEW3-tYO}IYS_pRk)ezJXzJ83~QUEWG#tWUgI|Y+~KtKA(-8^xHt~+Aiw%OIQ=5A+t
z&4`RoP)`3~o{X)$WiY2R;b(zNG2Br)zm>gqv5T1iu!8-N4c9l`J1qz4l6n}nlhD7P
zf9a}>T0}ij<#H$<cR|Y!?0hlHxxmjeAqRFy;R~girhmQ4GobI6M!e4U%Rq()?ZYzv
zO+8_~8A|9vJdsFukfgW_!mC<hc_;adR8c|<<03U1u<vtDP1pV1uUPXrA<K%3JH|f0
znR>jTHBAn`stnJ{TOOwlk{2as=N&4S`Yx_i_xTe%0SJGuWT(EV1)>e*mQu16B?kld
zS@47k*BX+VzL_U;WL{#a$u9giWZUu^I->v;P9UwtWhn2&jHg*|jfKh8h??)&+4&WQ
zUoB6+z#<FkWZ{kp4e0^2aPxpo+x8%4U1+%jM*|r8sFBZ5<ltbk6VVB+t|<t?KBV+l
zI+5p8NK<jWyP(i|d03u}ih9T#fVt58Y%NEcpP%0uTCNod-n{R1RH>L+DUFw|OYyk7
zv%b%fNo@1-X7`T<Gm=eWXyIBh;(%&aH(Ok(ZV`<ofiu&FYmX1KhJo<B>tvsFSpf1J
zQ;$ZCJ7~&oUzoKzs;-$LE^-jmC1+$7$*~sbU1=^5=I<YA7yHmQ{HJ54#G$rC8blxI
zL3e)Dab;ovY!(<%*5?-`Je-K5&ztoYO3F2aysLA**Z4DxUj`t`Ndl$8e%!9H1J{c&
z>$ze2nauMBNI`=MBPPvaDt7Jivh7l*2=x*5=FhpF%S5|1{w3Xm0o7C{US9&|BSq@V
zE-`;rJ*CP)MLz$K5QUrdM7YM_TSd>!8T*fw%Z=~3z>WpiG>vq^rgysk)CDhj(3hoK
zW^%Q>BQnZ*cR%P?OCzt>hpv=`M~5_vqYIrD&Ym&w+c=M|47CwiGL9eF8o9=L-3!b8
zj>zhYz~HbycXTR+*|dSW83`U*Mu9?{s^BVsj=yw^eNUzs`gTW*Kfc<&f{z}i)f23g
zEz&o5P5!f7iII&f(kQAGW(M|%Q$!Qf!^b3V$(0of@ept3U`s-{tiquU4veoxuH9-=
z<ZhmYrZ5*6f|<Y022YCMY*1)twu$C5?@w#kgXh1j5*foY$%B%0jmY;Xkq;ukvU%Md
zTD*A;O*cLBndc9Luf%!c<96d>t-Q=3kZ+mY0ZK(uvVu*jcprO!0_@d){?sfHxjRJd
zQB7Op>Vm%50D_e)q2H$L24ep!u>{4|PAlyK&uno)*4E=7YhPWe5@gayxj4ZDr%_yh
zaPWixWExFT(4T2jVwA;=ck6S9YgbrRwlycvIK>N<GhLuzRw*hfTe(`|6QlVVRB!eI
z9?#z;Q8294>dj%8Sd=i5q=#tD&2rG>D#Ctivz_9&$b4)4y|vi^2lnY39jQI)BKu9l
z_S(Z?up44J>Y+%$^PuYrSu|%5FmU-dkc{d_aAo0yj}f(BawU0Q2CkL@bnBLjSgpBS
zab>D4QjIgTz7^c)k+rt1nOIC@gep0Lp5*uH!I&_~tXHrB9jK=ehJOJi^&-+IK^SGL
z-It+Z&;km-Sm^{KnOO;6&krwMBD}#byJVobITIk*=gQtsjE+4bi=<+HW!7{m@XsL%
zV@fv#-Td+hDy35apw3Y6j__9KDYa?ci7r&isQ=<lt~nS<Pi<yW7g4KKnz+98q~1g?
zW$Z1Z+XL<=YTN9{Qfm&bEQMSZZfMKyCh)lYp_9n?%Ic0Yi3&(f;PS0`UZ>Ec-9upi
zW}n;cyhmOI)!Ac?Rpt730%Y?9)5O;VCqiF$6~TfTnW2Qe8PR9~s!$R@pmWJKn%X4T
zypj=OJ=t2E=fQL3dE9z%X1aV~Uo8(}0Sh=>VS*=zCAR2~#UZBLtv~Zt76U2zh^c-c
zWeMwlk*}B4AWMHn6x61@YGb0*^<k5Pvf6Z5cY`!Wn;w|*q{IijFV9ImBJlXiozenl
zg+fi^*oWfG#Ti?&MlTplrhpWwsfAB30g&zhR{5)8TTDwr%qhEh8#zL4@3drtF8CEg
zR!WGICB1<B%%AJH<LjRNNpXv&w)IG$lD?g(F%F!Vzi(|9h=}TIJ&Dh0+(XSwOz!Ej
z?b>%aa~8CA!0jwnDf-x*RAjKEzohE>8s4(|H-)EJ;Fkml*7HF(u%R^~Ue-^W7_nH_
zcKXu=<VW+N*e-t3x7(V(fgUp4-nIU0i^Y{UJ@@J!=Au$mW=Q{-^MT$I@Rk8J=S-G|
zGW4%Ju~01>)t6k9^_yo79|qJfLS)Xbq3Q(Jq_bJ}Hm|P2CI)-*ZSi-`8atp@#-J(c
zOqf{gcZdn_D+W>}sB$QuD$AXab0nr15b#=VeTv~2c&knF8edL+fUWDI$i>jiZGb=!
z<8~RcdGlESicLipC)XWxLyH^*k`veJj)o+ThgyMycw5?CmJoj(9qsqJiuC>rfemC0
zLk!*R*RMSug~OuOTmzGIbGV;`+{|as5yGUvVmC|GaJp$;tzO>GxCPFX%(RB6odq=$
zHGXm&$*A0;3%}_j>++R&(d~EDLS8`Wd~FTcyyol&j-*3Q-G0TD&aX@3zRW244bLr^
z6{TmRZhZ|e8|v}<rP5}T@@IY!ZF0EU+elZ#;d!I1lVVi!?~i|_T^2|d<tPGfU)!XM
z&JcbI*%aQXzZpp>dzayZ>Aczx60XeCF>Yc^;hw)kqvFS~h_743%xR`~&kVkm9h@V$
z5S)to*?JZas4^*c8BOwX<^yBUO6G(LnMU&{eA)wvOnM`HQtB}{#E^kuPiGH|Q!}-`
zkxvUZV_98{H*x|bl>dZiFwfurDa)-q@nl)#>}c3h&ynB_O|MJ{bYidFZbul{(-4wN
z19s)_5}oD>Jk+6oe-aBN2)^vh;HRs67z~cy+=DL@c9f+^$VJ4cS1n>DP@$;~jy_>j
z#GU!gU^z3=Y!(1W^_AtGD#FCb`?Vr@Ynwmj^JRRBOLzf_P5!XOXrY@8lW4h^eGEjL
z8H{&hZxf1eBX4iIAG<3hA2u!O+V+YKszzYxm5jA6i)ycJsF8{+2`{&8Zy9A7uLzg0
zXSzL*A9_U8QVWWe5)Whp+go7Jc?-51N0Vi&{|PMv&9FWU>ShL&I{}T$0uSjIUV(~#
zTPG8X{FAvT0HlLA)ZPIGH-A*m0&2sl4}+Zg-6vnh_;c_6nlA8MvV_vNZlzMqVX@a`
zeA%gJDjTU@VUC}7L8&rb4&4%QnRI862-ymlaCw>mfrBtIJR<qabcB*T&SCd4%e!yr
zG!80(nbS)K-(M5LK$=brg_-pIf~vcfoZ=xp8Gb7*hEd!?V4LFvwYvc-`jtZTGxXF#
zD&fhZ;6l!v#!yk)AgozVdx0R#K%_Y*NQtof`vJGOaxsD5bhr?NzCG?^<;{Vw&LVdk
zQ|N~I%8_p0i<ZHrAc3MNpgfd}SEAHr9`pO>O5T!|dUa0E6FBY#0AArQMG~o#hNQN9
zdvE9<uK8G$hBdqc`9*GtF+m+-G}!M6c_B94er7#&bfT(cPyr!+J87%^!d=S`&7H+o
zlvdWB5cMGJm@b6v9k?$Azw&A8?f$7M<@$(b?gI`^RG%As{(YERHdKOcfbR$@`uUQA
zK#QzDX<%f~<Yr0`AsPBgmf3<-r!YG13P}0$EIRygm{tLVv&1|kMcJuMl$15^sj8q;
z16t@zR4@=q1o8$OZWkVKHwbamW#er6a37FSGpJ^U+=7sVB$tEPm!jJ*BO-MYKb-|1
z>Gh_$$sfc?*`*@NshcX<J-BsXUC>|ZL`<%aq7un<(kO&1n?t=ua(0Qr`K+ot*{|?e
zqrQ1`%uqAwG#9M-{HD=pb{@vOZ)MleCycpw>j})nR-DS@#&X4v^*E$0QKFnC@!*0l
z4>#C40r@h?qoJZ0gI|t}+c_XGQI%Dcr!rQZFZik{Gj}t=kv(XMNAMR`@(X~la`VTL
z7S<2D5Bi;}ck0hTLNoS^CX@5$?PeGkd`}%&9b4o%$bm;%fW3Ib0CC;G%IZyem_+Hu
zMz_`Fpc19YL?4Wl*>^NzdDqzd4vi>8M)3*z$nIBsik5OuYCz-kbMy(&2jJCC@hr!k
z?fuIkVc0!r$zQ8N93QatL2;Wo{TdZE)ZkIGVG1{RX4hBe#oxO(V*Hby4~&Z;k@eC5
z27uB|gK$$OLFi9ziZ^#cgf7mh|Gj}?qTB5*u0o#yI^}-dEX0avJsn9sJ);v{Lr`ZA
zR-y=PB)5pc?SSZ4gXmYSA}`*B{TwkaeF9d@8t-iSQJ}^`eBg&4f$E`oJe~C`O42rr
z7y?9)bo};jEY|BWZZPI2g-*EpWv@^C_)|Yatuc;Him!30XW>8XV6QxHU!;WAWRAhu
zC&Un^Qn8eKwpZGgFWYAhRhw=7c~ffve#79D23w%#lxE!QkYt2$T5Vdu#p(A<2c<Lf
zBon2Obitdu=KOA>XZN}nrb)83k*2U9L@Pk0u9$C&@^39U2736nwi<K?w}Z}{EA<9g
zVWvDd>Q^38fw<|r5fdIG!xAGg2e-s2=fJ1iPZA+s=@&|Iqp99c5ChkY82zF2AJBi^
zRJsU+44Za$h0E0YqqUCyeA{Mt#-L@?mv^(6N0}Qg+Be9yye=R{=9>LH6Ll8``l<Fc
zWA;;s0{_Tr;zUCpa#m3Et_lRbKV5e*0a^kpMY}_v&XpIQWZIpSm}^Os5s6bggGh?>
z%-rHde(UmMB;PAO*m-Y9wlHHaQNmXo%x~aDt?)PF|LDrQ@!voiLoI&`1(t%irvEXj
z|1H?w{L@Vy)%hQP^9Zt4kx3t=-;0P`GX9hgiT*ZH&UgO;j>9sNBB2V33zy#&smu)&
z9L)Z$>uVBOezbwRAZ>IlA_DJ~9qqX7bz78j60)Ug#KVA|{I;PTPV$}2`))Rz^D4Uh
z<1yl1X-Lc}IdS{dxmnht5s#VyepWeMZJnpP3JVMcHWs9wD8%dm$xO6q+n(Ypx;Mw7
zq}4-KILL7oF*>cvK{<_&oYa|@&;JF4W%+!5c}GJyotFy`o`mLGnrS?EjD(#GM5i?m
z6V{6;TQ!Q+g?Pbu<5g|wnzV>Hpx-k56FCy?1eOx#5m^J{abxzOXH?^Ak7|eRb`iJ=
zW8)Gt8pYFdn18f{X;WVi6)M8vJnmFkWKN*H61tbWpfbQN$S$m&Dv2_xYLwaoM0894
z95+rP;scbUi90W}7CWziGv3I)`PJ>7pjJb}nN%gs!-<PgAi}_9?a&W1+ocaZn^Sgb
z0WmJ~Oip%^iYG_lhS3_`xsfzztG3doeC9n}*Z~FErnPB>)#YMc*YcdpN&yY()Jvx>
ziY9g1ivc@;D?V>?xXGrYlPDO53){VK9c=y0b19>byO*f<1uwx@+n9kwKE0vBUAK9I
zifFwR@s8ibOI_pR8na2H(l5Mi-K*xT=k(T^T-UHIw2!0djCK*Gas1ZNBw^{6x-4=Q
z`D3F+IfV^+$UyTIp``?ROBGupEV<ueMx`g&=Q*7p&+WTGh3a(aXW|JZTNu8lq^D+s
zT=NTiPKQ<i*OnyT`siw#Xb|hW>`7;FxHBY*uc2QXfw8WiRMSoQZ<W$kiVQ}-9t{jt
zIGj4=&iP97Lp;Cn+_-jdk11IIIVR<5?s%DVJ=aVLf-RPqlo>x`45i{u$ZG?R<AyBf
zZ^edLZm9G7U1L?M7t1(|bBv#`yC)XYR*jS;%zbrt)`^(-X=MoK`|PMyO3Y2XKb5!@
z&-i7RsNumcguR9A;~~=L9^SpKS;ac63I*K@!aZP<i@9ihFL;Hhj{bn2N1__N1EQVs
zAtUDG^UM_Ji!;?0gC98HgHHwGJjnubN1a=v{O7E5Q_mC>{uj0Re;n@~!e6QD`&Ebw
z&USD5TA@4|z2t_HR@8*EV*n=@g?%x4HyR>J!X#Q_pHVL!TKBzVkfCt=oxi*W=(Cx0
z^ez@`*JVf_hU-Y*?gx+&b8T3g)Qi=cYiI~Wcr{Xz<tuN|F{TLCF^W`p_NwPS`#|#|
z4uE}!eUo0A{F!E4BAks^lvZTz5+c8}s~t^5{Ent#aF~rk_>8m`=2ut}mkE)5psp#z
z6D<+<&7HM@{+OX~VRV)93mI?a#!AvyK=Y~ZxX4vM-?;5y8J5s_;YBc|W;s3fZOK?U
zOVl9T4q(~OW5x`JsmV?^ej0a2D$_$;Wr&0lg7k!Ih63yw&pO3u!WZXT>5aT2Koky<
z%F30&HQm&qz2`B;oX3`LK}OGk*_n2Iz6vOFQ;x>wk34U_K^B_Gq>6s}I}XT)F1I91
z%4s5ePM#Ag8KZACt?MmdZh7Zy_j>G{=QEiPCR<}IaieLWPh@^#dGv<)w-~T`<}M-x
zoqp*%LA?V-#;|IFE2a>J&(3;6zE-d*|M(lQJ5T%A=3<;&B30kEtyN<Vgl>)L{|n@6
zOVoTw!j7t2JZ;TAZpsD(h8!dKY?<!HVS+{48WD_*I3(;_zrm~t{Jr1L1yu=S>XE~O
zCLwGiPi%cZsxvOXIQD3a55C_N-YyK%I57-Iy`8ogNxU=+hA{RP&>}Mg(pp`cN6_>V
zqFULbcVUxc8ez!&!q{=RpDgU&<t9Q!c7mJ!w^z6{GvsMjX=Qbcrro~iw`Inxh-i}`
zjtuEDWPApNhtmcOZFt2NK(0Wo|1RDtU^q}b7qb_z&<_2V3**D@`l^?YqSbE6HEH;z
zsVgA*c+0}R=Fd_tra<&{mwzph65kR0M1(##+6*k#6lsrbZqdW5F3QG~@Gf>zD@bF;
zPizp2FFEl)HxLEycFMyMor>CCF~b^r^f%GaTe~t1VvUyosf2#R(QM*5yMcDONuLb6
z=?mTBv<`|_+idNJ`K-~F!B0z@5<QA|+w74Ecm039io5b$RSvxoVAB29Y5$+Jbfx=2
zgBAwW{VKOJugo6moee?WMs?qFRsiRur8(yG`_&b=%(4GkaIH7C%NilQ(9=4F(nS_k
zsRd&BPxqn<z`21?o<kW&BUNrwUc`tZ?7@*z1{69NU2Akq?r(_|DP(j+&&f|lXuKI6
zf`ywP8P{ajw;MBV(d7Y6Rl=5)uz+fnb)^;mvG~ZihAQ)MjSH)Eido|`uqf)!n1d3y
z1Zo;0@$BnA$@N&iD!^c@XZhyJRWB#(@B19w2%fuj>wO}T-~>w0Ek=tN%zZcuF7PpJ
zGf4ZYJFVJ=yYzdWTYuIz$Owmn&}Il>v^FCBEx}!!qG-u;7Fd{s2=xm#PUePGYb7P9
zpI%?&u#sWbG4AO50_h16BXGOs<;c7BmpPvA)6Y^;AXFmnF8?JmH%Lkkj?15^jp~xE
z>hE^1xs?~3GrqOe$U=d7UMcr0GdYF|k`5EuiHd19OAuQA-uKyLVN@flBw2r6F|%3Q
zmWV4Niym87sWAhb;^33~u2XH!6aPXwfWV!`s)ynqW@{I9G!-XDJTha-9RIC|-|D$V
z1B-pPhp-V&b?XkNkkFR~s{-lFC^?zp&Lzf7L2q?*=Y6yI9*WCY^{_XKU;J>ztVj|r
zUh%w74oReOb(Bzr#4;(GV40b>7&r7EWcthV5Gv21doQ>hd#g%+ZX6JW3u{BEi@kzJ
zM>{20pR$ewdCf+$;amhN=F2VcmvkeSyRGG>6RE^MS5R+wfL-@A(*7u=ki&`W8}zl)
zUS|=%3Gcd-@-tJ-#|&^xgd}+WRIW9$P#~Tb?Rt|b=mek(DplhzWGpU50P^`ki(N!*
zXmb((i3=tk`L_;uNolcGmRlmYd3bG)?4ekBYvdVSuB3Ynmtp-CTC_vL#%oP@`A@YX
z*1n)LjlIMns`qQfqW4P+Rbmq0_{0ENxxq*&e<;fkeQOQ@fhG9eH2F37tx#>q6JM76
znwrTo50d{0e_6lF0=7#kO{=rg4WN03WzMgX4$E%#s2=%Z%=tedazM5C;J0S=tcEOX
zz}U4<RR-H-3SXPbc@4ZR+RinCUmU8{qhEC%6F6q?XwUY_jMS=k_89IM@XpMWGWa3H
zJ^OaNXtEt7SU!Mx&U2h+DB@DA<p@g#O_bAU^jF+t!sj$Jez%#xBNj@Qa#g4hKhi*Z
zygARG$L|a)_b4}5q&6-tj*%tiSPvRBr@b*t*Z#&4e|0}_tZa+<6BNW-$dyXdM$h}Y
zSY;WdcR@;wN)`=>iuRy?gyN00_%7BvX169uw<b4XJN?#on)C@q(a5`)2$(hiNWiS9
z5V0JRfDcE?%n;XJpwW{XUQg(q&o{LBA;tNungP9}YsYd=OmJQcfAo#tF~I_t4fV4E
zR+Tiq<b4H{R0+1PLi`%u?+|Oi1brjNcz#gnQlVXkaA5SgMpcgZCy~7fyclM=h-@DW
z`+9={CL9h((sAnZqKoZ@SG@fUo6?$GQ1!tK{2OZP3_wOGSUu6kcm-@no-R{LgNuMU
zzPrll*Us18xe}svvJ^V^xw>31@SOWW8wMi|%PUq}vCw02Fi<Z>Kxj|%Lbx-dvCr3z
zY{V9(Mn(}yNl%(Ef^ur<Ed(LsxmfQM+hM7bhpz;8kkoplOOrHqD+&I#>P&W{xjRdZ
zQPdiDT&QI=dW38VGo@ngNv_0)2Fy+z;ukW?t_7+PieF7BD9=1Gn<By;OX|{^UhYx0
z#0A$yYJt+GtaYx%gi0}*p;g=FskJjrU7L$8EB%NiX92vk%R*KO`(&>yzs@gkK(Sv{
z@f#F4D;)USzz?u0B%HA=W`pa|$zEHb;`mVT=Be-U2CXS%GZ@+%$$ug~%n}`2;%Rcn
zc5tnd(gM8;NMI{W=ucTyOmT>ycpLZNzN|6X;bAX7surOlRwBJeu6s2Nn-OFsB}1{o
z*^M8!4<;jJd`q^;5eV~ORv6Wz*W@qHPMM-N3c~^oxqxh<*#GBq;<!1FNm-I;PaAhw
zcE~~ivObC0#K@=AHY*w&lTAU$pPKi7_N@r+js%960Fz=x)5?eC!N^)_qu^m@SE^vd
zP;h{Cfrwg>l;yGrF$H(#lkEeoZ|vdPBMo;5so$$uCd`^4B9|-qArVH;j~?n<maOhe
z%(W}RLawo%i+E5F3H?4*Vfsb0h`%TaTRd9e{yBZ6m8J1gqa2BCFSj52(IMqjKr$eW
z57Q{Q@Wa3S<<pA6Obx8R+C)ho>nEp?($MrNE#F;8)t(ccgUx7uUTW=4;HnfRa5%rg
zAE`VI{}-1dghe)$PA}h+An7Fkvk~x40Gkebp}NJt-9O~jf^xlEcd`ofZy2kirE3Bk
zB`#sPLDxQl(pG~H5BF><I6@_Y^<0RQ53L}H#XWNO6ASq=+X<#hRrMh=fr^#@_cT_P
zH)Ad$<F3^?F2pDcv|gzvp2rja2Ai9y4;?r3wZ9oH=!Zxzm`gs#$b~|C>uUl7h|{Hm
zAgNUrJGGmFf|(su1xFXz`JXijPIIJJgQz4|aT+#NT5i<K3WNfo{^uXHY}Me<E1YLA
zx$drGB(3>M&|QTALGZP3m2<~wEM(|X(8tBxN-ja5H*b82t%9^xc`=2K1Q>?IFUJ^R
z<_JoAFZs?9@ISxP9g86hLsKFTOFM@4#=f}WgwUG=p8jA)sfJa;2g#0|t)sxUdelkE
zX%ls0W_M>-iN9#PH6IsG4%C|Wq-lMy3@2GN>i@0JUyH}v8Z(I<J|1I<<C|5t%gfl+
z;Wsl1OdZ@6e?3xI^G+gI9fhz@Ea;UG4;*hVzM;Dg8=1{>86`f+Cl;b;XzLwuarnz;
zfG)_on{E(>slE8aff0BO0)|S2-oK2*)sx7Y<;K0jWWg3T%&gw$p@75yV7L2j&sQ=o
z00V-Sjj3w9C9+EDSykc5vBG2c?=a)p*R|1^o<ThJJARQmdniQNW3zlh*}h1wYcr%d
zmI>D(I#r<*BFq!I8HzP<?KVj!`&=vv2k`cu77^g8<SXN&j|dB8Yi8>xCAv#)(i?AA
z?ztYZ{_6SQY37BOGrD1#$asB_;en9+3oO(;W9#6*&*`(Ef<`VE(RuH{(#fRn#dh^j
zQE&S7fITwm^@L(*WgR<zX1&%6<*8@3p{LeYN$WO4;C3*1hb^X&?AF19O%#|i!EuxI
zu138c$CsIyc;xaTXWDy<QD<Z}?wiE{YspC2`~+%(fzpR!4}Oj?92lX{mNw#BnwO0=
zIDCZVN_0<T?-GaJ7pMt-v^^q&_X;T1*31BU##PPG#suZNy!lo-nJt$<vJEae8w@5;
zLOpvo%xBU-B7LBIuDAoEmbm%WSsIAS?rJd50pm3e2sBMJc`9Y*tu*`+V>Sku@#!5~
z6L87Lzs3JD;ieu@Ys621^Eh@Ndfr{DiwQ~7<7R;h`TgWqx9)cWuI#kD?kV8v!o_2W
zkIdpciov%k5wc}xoy0<m-Hy3Wu-WRXW<byQKaXocXKrxt$QZ0}WNmjdjz{C`8zg@T
zK1MwsVZcO7$soq#kEW`qtUhM^Z4lA&BEZ|xdA`=aO=#;sIC$egT_cy%++~WrsIdnc
z-vRuRn<?89$zTWRh$&O+`O|*g<GY#WhW)A&v#hyd)}PWwI>9e^6@y*2?-efW=WYl;
z=j^uF;fYBR4{YJP-y`;Xx%2ym@n}bkm9u4)hZMg=ezD<Vx7jeQh*)(H3lT^rQd?r#
zD9?lbZy|hs?LG^LwBPtl3kh;>YCCMMStay9%ajs`Q}Vj_hH_IENmlmqMnb44Nl21(
zebkro3E>4W7l)+HH=XuhIDfA6Z5F!7`A+JIw-}RW+wa`9%{jUMNNN!9UVXRtF<?u%
zS3zc>5K+_p;AMx0V(PcYQw<2~G~3$?+4sS33wH>%fzIu6Om9W7skw26%*2(-0(s@j
zoYG$=;VH+V^{J2_skS_?O^1PuS)63VFz1WPxl`^+`OMVkBOjRB>HAF@m`Q<A<kT%!
z{hDg_f2}4b6%|q-&C_?L)D9xYoczhTNmRU?LUS=8k;eReOTvZf&XY3P>V=jWK|0ir
zN_OF-Y;$&-Pf&EpJWJ5_OrbTvq&Q?@Y{{?MN1LOj-omjs$&T3rP@~&H1@ztZKiQel
zY49wnY+J}{W!cDqJw%*-YPOfuv$ImOnsu*U54srL4XnKx)0EtmIbPfS$d~Zc3`KH$
z4Ii<C+G{2eMYZAv<W1A2dldnCkpa(0nvkhDQP@}T`@Nhy#^5fDqrJ>Y`ge#wLVuqa
zn4WgpA@bB#FN)jC=~yYqo>`FyN}hxU`|)C)<o%<^e3?=<UXfKAg|4l!s9551l}^+-
zhMkB;sAGNSZhY=`c|f(PnrNe{v82jK5Q>vi8|r4GB*Lemm8pvUJKq2;`&W{EZCaGG
zA?nWzdD?sQGTH1T`#nm%oxgyC&E(3prc(G6cj`=S3i_YnCE035-|3Rl1}!vHTQPX@
zV>FIdH(2A@*X+G)8gh@Prd|Gmib7xLEbEl%9S;(<i`+m-vgIj~W_gyf%ujt&kcvAC
zxptE0nH5<stu~ua_I!u4;o>QORy678Is2S%R!Rc)D-7n%t_J7iXm9QGg;(x(nsPhc
z3LqnIEuuyCZV?h|=W-Qe+;<-v?za&0-H--Oom5K$3wDyIfGdu8+!{AKj}%HXISd;;
zo3kXVa{#hnx@p{OpJH^5dkf1%TvIj5f!<Ct^m`)_YajF<c>s}(a6JX`foVOT&M|rt
zu%GQTps#b^yHrd)-U~ZsbR2~BzBtu-u2D){As6DVjg4N$L2CW}&^RVBI!Oj08Zog%
zZ)22~s%G<MR`T&n&<Z0Qmm6e?8^Ogp7=y2PxyAa}JBKG6(7V4*5InzR+2Mfr0}GsC
z*<@}@`SeYgifPF4Pn9Sfr{gO0AL6tP=)wONT^TUT+UY0}Y+h%-4E*)OKD*{kih8kn
zZOF74CPX}KLl(-@ELY3o&(|a`yMs+HFxbVo`N9n6#o4@o>;m{GrRLRO9G5nS9j6oj
zN$19ohn4Pb|L{;RnYfSRi&gRCHU^x($FrOL$+9YOJGPiDTHAohq+euJ*BUY?sLJE2
zvZ;g6Z5Hvx9MIlbbfsOntZ`id&wA_}S<t$$Bh-*X4o7G&y6%!;Yj;+NE+9v(O!8yn
zX{%EA=IA=Tqfz%x_Az7qPVQ08Jn4jtcDN*t>o_U6s=5^!#&aF1o_9Rcxn?eb+;cJc
zGj;yyOIR&{NjkBC^gilSM6{ojnppS@#wkN2<<p=BrA3TxP-;F$hjpqYdaCXBft3Aw
zX7Vbz4)E;KO?t-%vO|97)=9TWD8su*gsgE$I{`Nf5Q}KS<G+}t54?TSr>4h%du~^u
zD0TZRi|f>Th7TmU>?PbbMgRAoNIrM(-04T)QXALfeN3~_=GlP7S(>`9gJ&KN1zC7d
z9W}rGpxg6~k@q9j42CLY2e?rKPsM5bcCH-pmSOEVS;NLFRv1V0;*k-DNsE}7I_#x@
z^ZN2D#N01(@>#t9AI`|DwzsbIyYt=m29JJ+nqBT*@G2K0pyJ7bqTP1Nbu+xz;Tm1L
zmA~aYrRY7Qg0J1Pz^8Xp*BIpoX}=TWGQ9uWU_I$oze97ho$H|*Tn}IH&?VJv-=EFL
ze!2@Wzgtyb8~OW@=YGE_j>n=sNYZVFP>6jF2w_RB0F;XuP?ARjgLLy-9Z1ym{+z`f
zWbA>LcXP0H^mFtRB2h0ddtDn;t#`~ps`Py)e({xp$)Y+mseI?%O)Sx0qm9ZwC(%Q2
z%V1j4KV9<dETUT$Fe(4I4Uy$-XZ!`s@P6kUaZWg6@u<1V^EItIhyBgK&L$C@b@i@q
zpwPnc+49dgHqxR4yn5#0^hcYWM2}29;NzZ24P6#XMcV#6X`6hGy<R5FSjg2`Zj9SC
zSmv~!_F?k-DqvUqSZCsbO_me{(05eq6Cdcbpp9qk1(3ZIuhd1tp6*i{{vViM1(Tym
z{}nLwkbeWL#0bTYFu$PT!r5T+oJJgcGb?dt)7L<E>JcM&cL<7l9XI!*12%PZA2p*`
z8*Q4Y?J2_t-Xs+LY;%ZiBL@h7sx03U@jWg-EB}45G~a1GaUvDyHKzv<l()6=nuWU7
z8a_p?{DLs}86m8nl2KZ11_0ISoE*%v`!3M6AtcQ_hu}lAP|Q;C1ep%|>z0KJ9I)c6
z)%4c0chf1}3sRZQ9hW^BW0P(z{FLSm-@zk`BOwF6ZUF#mk-pC$uL-T5sT}K|_*=*?
z1$3LYf6}=+x3zdE_g9khI*<c@fvvwFDAuw+_cYa4VYkFP6?Ani0SE7W6NIRo^0xyo
z5_(8Ag0g!ea5n`U^{S4z++BPlnEu#s4TFr-N<6)mTX|$y)?!czg#YSKGk(v;#KS5>
z<nLRd!ebq&!_<G8L->Fp@*lnL@nYY`;v#KAtf|Xw-TE2ue@FE7N}j*jAaapXx(T#k
z^ELHZ5&7i?%dzXSxqsI?;Ml$JrNbjBCs9G6jp%jepQ+nQsW!hO4~sUxL4ht;1XkWj
zer$vMlFi!vs)qIP_pLWS_KK*#*E>V0wy7g9DNHh?-o?w;hcAx5JtceSvoK$7fB6ak
z2?9Ij+7edYjb9QD9&N>g+kU6h<c#V=jI-#&K2PApzR{H5#C~yBmtQzQQEGxv2Ofef
zfZcxa-5mRsJf>df?%qvfI_uMth{KR?8QyL;{yq%&HL}thDJ&g5nzriC8Tf(Gun(EF
z7lCuzo*>yoTt}cQ-~PN$5R9OP=DMdV@^Bs+hoJ6<Pcb+4`n4@JV@O>q8muAA$R0S~
zfUXVeEb2*yb<DBOB85Aw5yOf>{eA(D=e99Cofmm7t6KJMLfS>I_m|o(P-g0bpzqvk
zKCH(i@=K#sQPhw0uHiJHmq`X=NnR94)%o520=D48W`ioMElz`A80+z(Lh$#E-%_io
z8MW*mr;@eK(=o0k#5PE|HT}OLfZfOM<YrUqA(k>9#)Ci2RcoE&Jg5CQB&LD2EE<yZ
zky_i~h+!(d^B@+&j2y_Xz4}`}7UW57m3@CWF!b?@UUwHlGhl6ruQQ;5goOPuDLoju
zQdDt{<#Y{vyV9m^(Pp0|6tsMI<m+}m5E(Cc$BYyC9uVBUr`~>S@;ghGvIFjsyIGM6
zvuecnq$K1|Wnx3Yg7e0gI{BdY*|$RtR$PVZK^G^G$Lskjnv)QMoT4uTA9@IC9yTm)
zyQ;EYN<!UK7A8@<b|8;Gjg^BRUKM>D<9SS+yL#*y$XqcD`(o5B|CzrQsri_V_CC4U
z>|o|#C?Bc4-vI!%1RI5c7I=jaWqwNeE8M*0Pd=XyMG*1l*F6})j$V63QvPW)Vmp6!
zg-cp6fVF?q<D3yCbq}A^Hf{lW@vZHbk9E;HNU;I84Qs7#Lbu=#L135M>z_Dxf-hO|
zN-iNZ(23O#oBmU?{fd%jSyAmWnr<w>^Xl72xq8A?fg!OQ^=0;96|mN<H5eVgZhNqP
zR&wk0dGqH`d62|;=7Z;2FwdoJQyaiL8`-aNK{l@=N)`x!+ZA6$svoc2_uc?8cr}pw
zbOel`y6eV-d)0<K>L<IEc>Hzl@lE0!wNJn#5Fmw|nbXNV^R4Jc{@@w!&ht^q-k)cW
zBcZLhML@bQ;T0v4h}a7M6}U@f&X$xp#uq?Qx0r{#oPZO9=JVc3WiQp(T<_)G49vj-
zR}HGo?>B0Uk>gVj8z*0NvSW4nyg!qC{69~EQMV)SY3`^`*Lfg>^~NV^K6~T6NRvg0
zh(<PXvrabK|Nim+9%ETh&V~rEbQ*5?uI2w-<-dFNzYBrKFz~7Ge^>c`FN`>R&Wip2
z{{4R!9w-2}@>wT)<9Uh50K0~pGatf5X+h&L=Rq9lb9;QTw%7eDxa1M|MRqW-jQ`Ki
h|GR?xqe<06FzA_+f7(vM%oE^CSwTa-TGle`{{u)>k~{za

literal 0
HcmV?d00001

diff --git a/index.html b/index.html
index e3e6eeb..25ed71f 100644
--- a/index.html
+++ b/index.html
@@ -366,6 +366,7 @@
       </div>
       <div class="panel-body">
         <ul>
+            <li><a href="./python/reverse.html">Reverse an array</a></li>
             <li><a href="./python/design_own_libs.html">Advice for designing your own libraries</a></li>
             <li><a href="./python/ensure_dir.html">Safely creating a folder if it doesn't exist</a></li>
             <li><a href="./python/list_files_directory.html">List all files in a directory</a></li>
@@ -578,6 +579,7 @@
       </div>
       <div class="panel-body">
         <ul>
+            <li><a href="./linux/simple_bash_function.html">Create a simple bash function</a></li>
             <li><a href="./linux/bashrc.html">What's inside my .bashrc ?</a></li>
             <li><a href="./linux/linux_process_information.html">How to check Linux process information?</a></li>
             <li><a href="./linux/directory_size.html">Check the size of a directory</a></li>
@@ -644,7 +646,7 @@
     <div class="panel panel-default">
       <div class="panel-body">
         <ul>
-            <li>Coming soon.</li>
+            <li><a href="./mathematics/Number_edges_Complete_graph.html">Number of edges in a Complete graph</a></li>
         </ul>
       </div>
     </div>
@@ -670,7 +672,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/jupyter/clear_cell.html b/jupyter/clear_cell.html
index df1d807..d9894b5 100644
--- a/jupyter/clear_cell.html
+++ b/jupyter/clear_cell.html
@@ -149,7 +149,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/jupyter/remote_run_notebook.html b/jupyter/remote_run_notebook.html
index 2f0b9ad..042cbec 100644
--- a/jupyter/remote_run_notebook.html
+++ b/jupyter/remote_run_notebook.html
@@ -211,7 +211,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/Reloading_.bashrc.html b/linux/Reloading_.bashrc.html
index a36d0b0..193fa0a 100644
--- a/linux/Reloading_.bashrc.html
+++ b/linux/Reloading_.bashrc.html
@@ -144,7 +144,7 @@ However, you can use the following command to do it </p>
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/Reloading_fstab.html b/linux/Reloading_fstab.html
index 1e0a922..e673a7f 100644
--- a/linux/Reloading_fstab.html
+++ b/linux/Reloading_fstab.html
@@ -143,7 +143,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/bashrc.html b/linux/bashrc.html
index 820c30b..a6db542 100644
--- a/linux/bashrc.html
+++ b/linux/bashrc.html
@@ -203,7 +203,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/directory_size.html b/linux/directory_size.html
index a4f6356..9169de1 100644
--- a/linux/directory_size.html
+++ b/linux/directory_size.html
@@ -148,7 +148,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/free_disk_space_linux.html b/linux/free_disk_space_linux.html
index f878148..44cbae3 100644
--- a/linux/free_disk_space_linux.html
+++ b/linux/free_disk_space_linux.html
@@ -167,7 +167,7 @@ none               0     0     0    - /config
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/get_ip_linux.html b/linux/get_ip_linux.html
index 9a22785..d327ecd 100644
--- a/linux/get_ip_linux.html
+++ b/linux/get_ip_linux.html
@@ -256,7 +256,7 @@ wifi0     Link encap:UNSPEC  HWaddr 5C-51-4F-41-7A-AD-00-00-00-00-00-00-00-00-00
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/linux_process_information.html b/linux/linux_process_information.html
index 105ccbc..e191213 100644
--- a/linux/linux_process_information.html
+++ b/linux/linux_process_information.html
@@ -172,7 +172,7 @@ USER;PID;%CPU;%MEM;VSZ;RSS;TTY;STAT;START;TIME;COMMAND</p>
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/mount_nfs_share_fstab.html b/linux/mount_nfs_share_fstab.html
index cb085b0..1f22500 100644
--- a/linux/mount_nfs_share_fstab.html
+++ b/linux/mount_nfs_share_fstab.html
@@ -158,7 +158,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/share_nfs_share.html b/linux/share_nfs_share.html
index e5030d6..2864337 100644
--- a/linux/share_nfs_share.html
+++ b/linux/share_nfs_share.html
@@ -151,7 +151,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/linux/simple_bash_function.html b/linux/simple_bash_function.html
new file mode 100644
index 0000000..c976d38
--- /dev/null
+++ b/linux/simple_bash_function.html
@@ -0,0 +1,257 @@
+<!DOCTYPE html>
+<html lang="fr">
+
+<head>
+      <meta charset="utf-8">
+    <meta http-equiv="X-UA-Compatible" content="IE=edge">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <!-- The above 3 meta tags *must* come first in the head; any other head content must come *after* these tags -->
+    <meta name="description" content="Data Science for Political and Social Phenomena">
+    <meta name="author" content="Guillaume Redoulès">
+    <link rel="icon" href="../favicon.ico">
+
+    <title>Create a simple bash function - Linux</title>
+
+    <!-- JQuery -->
+    <script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.3/jquery.min.js"></script>
+    <script>
+        window.jQuery || document.write('<script src="../theme/js/jquery.min.js"><\/script>')
+    </script>
+
+    <!-- Bootstrap core CSS -->
+    <link rel="stylesheet" href="../theme/css/bootstrap.css" />
+    <!-- IE10 viewport hack for Surface/desktop Windows 8 bug -->
+    <link rel="stylesheet" type="text/css" href="../theme/css/ie10-viewport-bug-workaround.css" />
+    <!-- Custom styles for this template -->
+    <link rel="stylesheet" type="text/css" href="../theme/css/style.css" />
+    <link rel="stylesheet" type="text/css" href="../theme/css/notebooks.css" />
+    <link href='https://fonts.googleapis.com/css?family=PT+Serif:400,700|Roboto:400,500,700' rel='stylesheet' type='text/css'>
+
+    <!-- HTML5 shim and Respond.js for IE8 support of HTML5 elements and media queries -->
+    <!--[if lt IE 9]>
+      <script src="https://oss.maxcdn.com/html5shiv/3.7.2/html5shiv.min.js"></script>
+      <script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
+    <![endif]-->
+        
+
+    <meta name="tags" content="Bash" />
+
+
+</head>
+
+<body>
+
+    <div class="navbar navbar-fixed-top">
+        <div class="container">
+            <div class="navbar-header">
+                <button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".navbar-collapse">
+                    <span class="icon-bar"></span>
+                    <span class="icon-bar"></span>
+                    <span class="icon-bar"></span>
+                </button>
+                <a class="navbar-brand" href="..">Guillaume Redoulès</a>
+            </div>
+            <div class="navbar-collapse collapse" id="searchbar">
+
+                <ul class="nav navbar-nav navbar-right">
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">About<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="../pages/about.html">About Guillaume</a></li>
+                            <li><a href="https://github.com/redoules">GitHub</a></li>
+                            <li><a href="https://www.linkedin.com/in/guillaume-redoul%C3%A8s-33923860/">LinkedIn</a></li>
+                        </ul>
+                    </li>
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Data Science<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="..#Blog">Blog</a></li>
+                            <li><a href="..#Python">Python</a></li>
+                            <li><a href="..#Bash">Bash</a></li>
+                            <li><a href="..#SQL">SQL</a></li>
+                            <li><a href="..#Mathematics">Mathematics</a></li>
+                            <li><a href="..#Machine_Learning">Machine Learning</a></li>
+                            <li><a href="..#Projects">Projects</a></li>
+                        </ul>
+                    </li>
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Projects<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="https://github.com/redoules/redoules.github.io">Notes (Github)</a></li>
+                        </ul>
+                    </li>
+
+                    <!--<li class="dropdown">
+                        <a href="../feeds/blog.rss.xml">Blog RSS</a>
+                    </li>-->
+
+
+                </ul>
+
+                <form class="navbar-form" action="../search.html" onsubmit="return validateForm(this.elements['q'].value);">
+                    <div class="form-group" style="display:inline;">
+                        <div class="input-group" style="display:table;">
+                            <span class="input-group-addon" style="width:1%;"><span class="glyphicon glyphicon-search"></span></span>
+                            <input class="form-control search-query" name="q" id="tipue_search_input" placeholder="e.g. scikit KNN, pandas merge" required autocomplete="off" type="text">
+                        </div>
+                    </div>
+                </form>
+
+            </div>
+            <!--/.nav-collapse -->
+        </div>
+    </div>
+
+
+
+    <!-- end of header section -->
+    <div class="container">
+<!-- <div class="alert alert-warning" role="alert">
+    Did you find this page useful? Please do me a quick favor and <a href="#" class="alert-link">endorse me for data science on LinkedIn</a>.
+</div> -->
+<section id="content" class="body">
+    <header>
+    <h1>
+      Create a simple bash function
+    </h1>
+<ol class="breadcrumb">
+    <li>
+        <time class="published" datetime="2018-11-07T09:37:00+01:00">
+            07 novembre 2018
+        </time>
+    </li>
+    <li>Linux</li>
+    <li>Bash</li>
+</ol>
+</header>
+<div class='article_content'>
+<h2>A basic function</h2>
+<p>The synthaxe to define a function is :</p>
+<div class="highlight"><pre><span></span><span class="ch">#!/bin/bash</span>
+<span class="c1"># Basic function</span>
+my_function <span class="o">()</span> <span class="o">{</span>
+<span class="nb">echo</span> Text displayed by my_function
+<span class="o">}</span>
+
+<span class="c1">#once defined, you can use it like so :</span>
+my_function
+</pre></div>
+
+
+<p>and it should return </p>
+<div class="highlight"><pre><span></span>user@bash : ./my_function.sh
+Text displayed by my_function
+</pre></div>
+
+
+<h2>Function with arguments</h2>
+<p>When used, the arguments are specified directly after the function name. Whithin the function they are accessible this the $ symbol followed by the number of the arguement.</p>
+<p>Hence <code>$1</code> will take the value of the first arguement, <code>$2</code> will take the value of the second arguement and so on.</p>
+<div class="highlight"><pre><span></span><span class="ch">#!/bin/bash</span>
+<span class="c1"># Passing arguments to a function</span>
+say_hello <span class="o">()</span> <span class="o">{</span>
+<span class="nb">echo</span> Hello <span class="nv">$1</span>
+<span class="o">}</span>
+say_hello Guillaume
+</pre></div>
+
+
+<p>and it should return </p>
+<div class="highlight"><pre><span></span>user@bash : ./function_arguements.sh
+Hello Guillaume
+</pre></div>
+
+
+<h2>Overriding Commands</h2>
+<p>Using the previous example, let's override the echo function in order to make it say hello. To do so, you just need to name the function with the same name as the command you want to replace. When you are calling the original function, make sure you are using the <code>builtin</code> keyword  </p>
+<div class="highlight"><pre><span></span><span class="ch">#!/bin/bash</span>
+<span class="c1"># Overriding a function</span>
+<span class="nb">echo</span> <span class="o">()</span> <span class="o">{</span>
+<span class="nb">builtin</span> <span class="nb">echo</span> Hello <span class="nv">$1</span>
+<span class="o">}</span>
+<span class="nb">echo</span> Guillaume
+</pre></div>
+
+
+<div class="highlight"><pre><span></span>user@bash : ./function_arguements.sh
+Hello Guillaume
+</pre></div>
+
+
+<h2>Returning values</h2>
+<p>Use the keyword return to send back a value to the main program. The returned value will be stored in the <code>$?</code> variable</p>
+<div class="highlight"><pre><span></span><span class="ch">#!/bin/bash</span>
+<span class="c1"># Retruning a value </span>
+secret_number <span class="o">()</span> <span class="o">{</span>
+<span class="k">return</span> <span class="m">126</span>
+<span class="o">}</span>
+secret_number
+<span class="nb">echo</span> The secret number is <span class="nv">$?</span>
+</pre></div>
+
+
+<p>This code should return </p>
+<div class="highlight"><pre><span></span>user@bash : ./retrun_value.sh
+The secret number is 126
+</pre></div>
+</div>
+    <aside>
+    <div class="bug-reporting__panel">
+        <h3>Find an error or bug? Have a suggestion?</h3>
+        <p>Everything on this site is avaliable on GitHub. Head on over and <a href='https://github.com/redoules/redoules.github.io/issues/new'>submit an issue.</a> You can also message me directly by <a href='mailto:guillaume.redoules@gadz.org'>email</a>.</p>
+    </div>
+    </aside>
+</section>
+
+    </div>
+    <!-- start of footer section -->
+    <footer class="footer">
+        <div class="container">
+            <p class="text-muted">
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <br/>
+                Copyright &copy; Guillaume Redoulès,
+                    <time datetime="2018">2018</time>.
+                </center>
+            </p>
+        </div>
+    </footer>
+
+    <!-- This jQuery line finds any span that contains code highlighting classes and then selects the parent <pre> tag and adds a border. This is done as a workaround to visually distinguish the code inputs and outputs -->
+    <script>
+        $( ".hll, .n, .c, .err, .k, .o, .cm, .cp, .c1, .cs, .gd, .ge, .gr, .gh, .gi, .go, .gp, .gs, .gu, .gt, .kc, .kd, .kn, .kp, .kr, .kt, .m, .s, .na, .nb, .nc, .no, .nd, .ni, .ne, .nf, .nl, .nn, .nt, .nv, .ow, .w, .mf, .mh, .mi, .mo, .sb, .sc, .sd, .s2, .se, .sh, .si, .sx, .sr, .s1, .ss, .bp, .vc, .vg, .vi, .il" ).parent( "pre" ).css( "border", "1px solid #DEDEDE" );
+    </script>
+
+    
+    <!-- Load Google Analytics -->
+    <script>
+        /*
+        (function(i, s, o, g, r, a, m) {
+            i['GoogleAnalyticsObject'] = r;
+            i[r] = i[r] || function() {
+                (i[r].q = i[r].q || []).push(arguments)
+            }, i[r].l = 1 * new Date();
+            a = s.createElement(o),
+                m = s.getElementsByTagName(o)[0];
+            a.async = 1;
+            a.src = g;
+            m.parentNode.insertBefore(a, m)
+        })(window, document, 'script', '//www.google-analytics.com/analytics.js', 'ga');
+
+        ga('create', 'UA-66582-32', 'auto');
+        ga('send', 'pageview');
+        */
+    </script>
+    <!-- End of Google Analytics -->
+
+    <!-- Bootstrap core JavaScript
+      ================================================== -->
+    <!-- Placed at the end of the document so the pages load faster -->
+    <script src="../theme/js/bootstrap.min.js"></script>
+    <!-- IE10 viewport hack for Surface/desktop Windows 8 bug -->
+    <script src="../theme/js/ie10-viewport-bug-workaround.js"></script>
+
+
+</body>
+
+</html>
\ No newline at end of file
diff --git a/linux/version_kernel.html b/linux/version_kernel.html
index 7c59731..8ab2f36 100644
--- a/linux/version_kernel.html
+++ b/linux/version_kernel.html
@@ -160,7 +160,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/machine-learning/Source code for the recommandation engine for articles.html b/machine-learning/Source code for the recommandation engine for articles.html
index caed0f4..19fde38 100644
--- a/machine-learning/Source code for the recommandation engine for articles.html	
+++ b/machine-learning/Source code for the recommandation engine for articles.html	
@@ -1033,7 +1033,7 @@ Top dislike features:
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/mathematics/Number_edges_Complete_graph.html b/mathematics/Number_edges_Complete_graph.html
new file mode 100644
index 0000000..573cc3b
--- /dev/null
+++ b/mathematics/Number_edges_Complete_graph.html
@@ -0,0 +1,190 @@
+<!DOCTYPE html>
+<html lang="fr">
+
+<head>
+      <meta charset="utf-8">
+    <meta http-equiv="X-UA-Compatible" content="IE=edge">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <!-- The above 3 meta tags *must* come first in the head; any other head content must come *after* these tags -->
+    <meta name="description" content="Data Science for Political and Social Phenomena">
+    <meta name="author" content="Guillaume Redoulès">
+    <link rel="icon" href="../favicon.ico">
+
+    <title>Number of edges in a Complete graph - Mathematics</title>
+
+    <!-- JQuery -->
+    <script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.3/jquery.min.js"></script>
+    <script>
+        window.jQuery || document.write('<script src="../theme/js/jquery.min.js"><\/script>')
+    </script>
+
+    <!-- Bootstrap core CSS -->
+    <link rel="stylesheet" href="../theme/css/bootstrap.css" />
+    <!-- IE10 viewport hack for Surface/desktop Windows 8 bug -->
+    <link rel="stylesheet" type="text/css" href="../theme/css/ie10-viewport-bug-workaround.css" />
+    <!-- Custom styles for this template -->
+    <link rel="stylesheet" type="text/css" href="../theme/css/style.css" />
+    <link rel="stylesheet" type="text/css" href="../theme/css/notebooks.css" />
+    <link href='https://fonts.googleapis.com/css?family=PT+Serif:400,700|Roboto:400,500,700' rel='stylesheet' type='text/css'>
+
+    <!-- HTML5 shim and Respond.js for IE8 support of HTML5 elements and media queries -->
+    <!--[if lt IE 9]>
+      <script src="https://oss.maxcdn.com/html5shiv/3.7.2/html5shiv.min.js"></script>
+      <script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
+    <![endif]-->
+        
+
+    <meta name="tags" content="Basics" />
+
+
+</head>
+
+<body>
+
+    <div class="navbar navbar-fixed-top">
+        <div class="container">
+            <div class="navbar-header">
+                <button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".navbar-collapse">
+                    <span class="icon-bar"></span>
+                    <span class="icon-bar"></span>
+                    <span class="icon-bar"></span>
+                </button>
+                <a class="navbar-brand" href="..">Guillaume Redoulès</a>
+            </div>
+            <div class="navbar-collapse collapse" id="searchbar">
+
+                <ul class="nav navbar-nav navbar-right">
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">About<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="../pages/about.html">About Guillaume</a></li>
+                            <li><a href="https://github.com/redoules">GitHub</a></li>
+                            <li><a href="https://www.linkedin.com/in/guillaume-redoul%C3%A8s-33923860/">LinkedIn</a></li>
+                        </ul>
+                    </li>
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Data Science<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="..#Blog">Blog</a></li>
+                            <li><a href="..#Python">Python</a></li>
+                            <li><a href="..#Bash">Bash</a></li>
+                            <li><a href="..#SQL">SQL</a></li>
+                            <li><a href="..#Mathematics">Mathematics</a></li>
+                            <li><a href="..#Machine_Learning">Machine Learning</a></li>
+                            <li><a href="..#Projects">Projects</a></li>
+                        </ul>
+                    </li>
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Projects<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="https://github.com/redoules/redoules.github.io">Notes (Github)</a></li>
+                        </ul>
+                    </li>
+
+                    <!--<li class="dropdown">
+                        <a href="../feeds/blog.rss.xml">Blog RSS</a>
+                    </li>-->
+
+
+                </ul>
+
+                <form class="navbar-form" action="../search.html" onsubmit="return validateForm(this.elements['q'].value);">
+                    <div class="form-group" style="display:inline;">
+                        <div class="input-group" style="display:table;">
+                            <span class="input-group-addon" style="width:1%;"><span class="glyphicon glyphicon-search"></span></span>
+                            <input class="form-control search-query" name="q" id="tipue_search_input" placeholder="e.g. scikit KNN, pandas merge" required autocomplete="off" type="text">
+                        </div>
+                    </div>
+                </form>
+
+            </div>
+            <!--/.nav-collapse -->
+        </div>
+    </div>
+
+
+
+    <!-- end of header section -->
+    <div class="container">
+<!-- <div class="alert alert-warning" role="alert">
+    Did you find this page useful? Please do me a quick favor and <a href="#" class="alert-link">endorse me for data science on LinkedIn</a>.
+</div> -->
+<section id="content" class="body">
+    <header>
+    <h1>
+      Number of edges in a Complete graph
+    </h1>
+<ol class="breadcrumb">
+    <li>
+        <time class="published" datetime="2018-11-07T09:23:00+01:00">
+            07 novembre 2018
+        </time>
+    </li>
+    <li>Mathematics</li>
+    <li>Basics</li>
+</ol>
+</header>
+<div class='article_content'>
+<p>A complete graph contains $\frac{n(n-1)}{2}$ edges where $n$ is the number of vertices (or nodes).</p>
+<p><img alt="Complete Graph" src="../images/Number_of_edges_in_a_Complete_graph/complete_graph.png"></p>
+</div>
+    <aside>
+    <div class="bug-reporting__panel">
+        <h3>Find an error or bug? Have a suggestion?</h3>
+        <p>Everything on this site is avaliable on GitHub. Head on over and <a href='https://github.com/redoules/redoules.github.io/issues/new'>submit an issue.</a> You can also message me directly by <a href='mailto:guillaume.redoules@gadz.org'>email</a>.</p>
+    </div>
+    </aside>
+</section>
+
+    </div>
+    <!-- start of footer section -->
+    <footer class="footer">
+        <div class="container">
+            <p class="text-muted">
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <br/>
+                Copyright &copy; Guillaume Redoulès,
+                    <time datetime="2018">2018</time>.
+                </center>
+            </p>
+        </div>
+    </footer>
+
+    <!-- This jQuery line finds any span that contains code highlighting classes and then selects the parent <pre> tag and adds a border. This is done as a workaround to visually distinguish the code inputs and outputs -->
+    <script>
+        $( ".hll, .n, .c, .err, .k, .o, .cm, .cp, .c1, .cs, .gd, .ge, .gr, .gh, .gi, .go, .gp, .gs, .gu, .gt, .kc, .kd, .kn, .kp, .kr, .kt, .m, .s, .na, .nb, .nc, .no, .nd, .ni, .ne, .nf, .nl, .nn, .nt, .nv, .ow, .w, .mf, .mh, .mi, .mo, .sb, .sc, .sd, .s2, .se, .sh, .si, .sx, .sr, .s1, .ss, .bp, .vc, .vg, .vi, .il" ).parent( "pre" ).css( "border", "1px solid #DEDEDE" );
+    </script>
+
+    
+    <!-- Load Google Analytics -->
+    <script>
+        /*
+        (function(i, s, o, g, r, a, m) {
+            i['GoogleAnalyticsObject'] = r;
+            i[r] = i[r] || function() {
+                (i[r].q = i[r].q || []).push(arguments)
+            }, i[r].l = 1 * new Date();
+            a = s.createElement(o),
+                m = s.getElementsByTagName(o)[0];
+            a.async = 1;
+            a.src = g;
+            m.parentNode.insertBefore(a, m)
+        })(window, document, 'script', '//www.google-analytics.com/analytics.js', 'ga');
+
+        ga('create', 'UA-66582-32', 'auto');
+        ga('send', 'pageview');
+        */
+    </script>
+    <!-- End of Google Analytics -->
+
+    <!-- Bootstrap core JavaScript
+      ================================================== -->
+    <!-- Placed at the end of the document so the pages load faster -->
+    <script src="../theme/js/bootstrap.min.js"></script>
+    <!-- IE10 viewport hack for Surface/desktop Windows 8 bug -->
+    <script src="../theme/js/ie10-viewport-bug-workaround.js"></script>
+
+
+</body>
+
+</html>
\ No newline at end of file
diff --git a/pages/about.html b/pages/about.html
index bf5eefa..311f2c3 100644
--- a/pages/about.html
+++ b/pages/about.html
@@ -145,7 +145,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Creating_a_sqlite_database.html b/python/Creating_a_sqlite_database.html
index 82e36c7..ac23deb 100644
--- a/python/Creating_a_sqlite_database.html
+++ b/python/Creating_a_sqlite_database.html
@@ -192,7 +192,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html b/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html
index 72f08b7..8197d84 100644
--- a/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html
+++ b/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html
@@ -241,7 +241,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Iterating_over_a_dataframe.html b/python/Iterating_over_a_dataframe.html
index 1e16ac4..b0e3e10 100644
--- a/python/Iterating_over_a_dataframe.html
+++ b/python/Iterating_over_a_dataframe.html
@@ -235,7 +235,7 @@ At line 5 there is a plum which is purple and contains 28 kcal
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Moving_average_pandas.html b/python/Moving_average_pandas.html
index b04dd35..e860eb6 100644
--- a/python/Moving_average_pandas.html
+++ b/python/Moving_average_pandas.html
@@ -378,7 +378,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Opening_SQLite_database.html b/python/Opening_SQLite_database.html
index f258b29..8bf3c6f 100644
--- a/python/Opening_SQLite_database.html
+++ b/python/Opening_SQLite_database.html
@@ -168,7 +168,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Opening_file.html b/python/Opening_file.html
index 5210f91..68f7136 100644
--- a/python/Opening_file.html
+++ b/python/Opening_file.html
@@ -192,7 +192,7 @@ If you try to open a non existing file in read mode 'r', a FileNotFoundError wil
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Optimized_numpy_random_intel.html b/python/Optimized_numpy_random_intel.html
index a3ba007..2fdde47 100644
--- a/python/Optimized_numpy_random_intel.html
+++ b/python/Optimized_numpy_random_intel.html
@@ -176,7 +176,7 @@ conda install numpy --channel intel --override-channels
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Reading_data_from_a_sql_database_with_pandas.html b/python/Reading_data_from_a_sql_database_with_pandas.html
index 2159aae..e08984d 100644
--- a/python/Reading_data_from_a_sql_database_with_pandas.html
+++ b/python/Reading_data_from_a_sql_database_with_pandas.html
@@ -237,7 +237,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Saving_a_matplotlib_figure_with_a_high_resolution.html b/python/Saving_a_matplotlib_figure_with_a_high_resolution.html
index 69c2855..7510331 100644
--- a/python/Saving_a_matplotlib_figure_with_a_high_resolution.html
+++ b/python/Saving_a_matplotlib_figure_with_a_high_resolution.html
@@ -186,7 +186,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html b/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html
index e3dc59d..bcc81ed 100644
--- a/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html
+++ b/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html
@@ -171,7 +171,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/Writing_data_to_a_sql_database_with_pandas.html b/python/Writing_data_to_a_sql_database_with_pandas.html
index d17de6c..a7260f8 100644
--- a/python/Writing_data_to_a_sql_database_with_pandas.html
+++ b/python/Writing_data_to_a_sql_database_with_pandas.html
@@ -339,7 +339,7 @@ However, exporting to a sql database can prove very useful. Indeed, having a wel
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/dask_distributed_parallelism.html b/python/dask_distributed_parallelism.html
index d656907..dfa804b 100644
--- a/python/dask_distributed_parallelism.html
+++ b/python/dask_distributed_parallelism.html
@@ -338,7 +338,7 @@ http://127.0.0.1:8787/status</p>
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/dask_infiniband.html b/python/dask_infiniband.html
index a75abf9..99faf2e 100644
--- a/python/dask_infiniband.html
+++ b/python/dask_infiniband.html
@@ -148,7 +148,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/design_own_libs.html b/python/design_own_libs.html
index 24a8417..8305a93 100644
--- a/python/design_own_libs.html
+++ b/python/design_own_libs.html
@@ -232,7 +232,7 @@ The list of dependencies required to test, build and generate the doc are listed
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/download_page.html b/python/download_page.html
index 2bc3eb9..ee95f7c 100644
--- a/python/download_page.html
+++ b/python/download_page.html
@@ -218,7 +218,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/ensure_dir.html b/python/ensure_dir.html
index a662dc4..8373602 100644
--- a/python/ensure_dir.html
+++ b/python/ensure_dir.html
@@ -177,7 +177,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/list_files_directory.html b/python/list_files_directory.html
index b9ac826..cb9007a 100644
--- a/python/list_files_directory.html
+++ b/python/list_files_directory.html
@@ -213,7 +213,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/logplot.html b/python/logplot.html
index 94c6fc5..bd469d1 100644
--- a/python/logplot.html
+++ b/python/logplot.html
@@ -248,7 +248,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/reverse.html b/python/reverse.html
new file mode 100644
index 0000000..bd1c8c7
--- /dev/null
+++ b/python/reverse.html
@@ -0,0 +1,222 @@
+<!DOCTYPE html>
+<html lang="fr">
+
+<head>
+      <meta charset="utf-8">
+    <meta http-equiv="X-UA-Compatible" content="IE=edge">
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <!-- The above 3 meta tags *must* come first in the head; any other head content must come *after* these tags -->
+    <meta name="description" content="Data Science for Political and Social Phenomena">
+    <meta name="author" content="Guillaume Redoulès">
+    <link rel="icon" href="../favicon.ico">
+
+    <title>Reverse an array - Python</title>
+
+    <!-- JQuery -->
+    <script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.3/jquery.min.js"></script>
+    <script>
+        window.jQuery || document.write('<script src="../theme/js/jquery.min.js"><\/script>')
+    </script>
+
+    <!-- Bootstrap core CSS -->
+    <link rel="stylesheet" href="../theme/css/bootstrap.css" />
+    <!-- IE10 viewport hack for Surface/desktop Windows 8 bug -->
+    <link rel="stylesheet" type="text/css" href="../theme/css/ie10-viewport-bug-workaround.css" />
+    <!-- Custom styles for this template -->
+    <link rel="stylesheet" type="text/css" href="../theme/css/style.css" />
+    <link rel="stylesheet" type="text/css" href="../theme/css/notebooks.css" />
+    <link href='https://fonts.googleapis.com/css?family=PT+Serif:400,700|Roboto:400,500,700' rel='stylesheet' type='text/css'>
+
+    <!-- HTML5 shim and Respond.js for IE8 support of HTML5 elements and media queries -->
+    <!--[if lt IE 9]>
+      <script src="https://oss.maxcdn.com/html5shiv/3.7.2/html5shiv.min.js"></script>
+      <script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
+    <![endif]-->
+        
+
+    <meta name="tags" content="Basics" />
+
+
+</head>
+
+<body>
+
+    <div class="navbar navbar-fixed-top">
+        <div class="container">
+            <div class="navbar-header">
+                <button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".navbar-collapse">
+                    <span class="icon-bar"></span>
+                    <span class="icon-bar"></span>
+                    <span class="icon-bar"></span>
+                </button>
+                <a class="navbar-brand" href="..">Guillaume Redoulès</a>
+            </div>
+            <div class="navbar-collapse collapse" id="searchbar">
+
+                <ul class="nav navbar-nav navbar-right">
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">About<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="../pages/about.html">About Guillaume</a></li>
+                            <li><a href="https://github.com/redoules">GitHub</a></li>
+                            <li><a href="https://www.linkedin.com/in/guillaume-redoul%C3%A8s-33923860/">LinkedIn</a></li>
+                        </ul>
+                    </li>
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Data Science<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="..#Blog">Blog</a></li>
+                            <li><a href="..#Python">Python</a></li>
+                            <li><a href="..#Bash">Bash</a></li>
+                            <li><a href="..#SQL">SQL</a></li>
+                            <li><a href="..#Mathematics">Mathematics</a></li>
+                            <li><a href="..#Machine_Learning">Machine Learning</a></li>
+                            <li><a href="..#Projects">Projects</a></li>
+                        </ul>
+                    </li>
+                    <li class="dropdown">
+                        <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="false">Projects<span class="caret"></span></a>
+                        <ul class="dropdown-menu">
+                            <li><a href="https://github.com/redoules/redoules.github.io">Notes (Github)</a></li>
+                        </ul>
+                    </li>
+
+                    <!--<li class="dropdown">
+                        <a href="../feeds/blog.rss.xml">Blog RSS</a>
+                    </li>-->
+
+
+                </ul>
+
+                <form class="navbar-form" action="../search.html" onsubmit="return validateForm(this.elements['q'].value);">
+                    <div class="form-group" style="display:inline;">
+                        <div class="input-group" style="display:table;">
+                            <span class="input-group-addon" style="width:1%;"><span class="glyphicon glyphicon-search"></span></span>
+                            <input class="form-control search-query" name="q" id="tipue_search_input" placeholder="e.g. scikit KNN, pandas merge" required autocomplete="off" type="text">
+                        </div>
+                    </div>
+                </form>
+
+            </div>
+            <!--/.nav-collapse -->
+        </div>
+    </div>
+
+
+
+    <!-- end of header section -->
+    <div class="container">
+<!-- <div class="alert alert-warning" role="alert">
+    Did you find this page useful? Please do me a quick favor and <a href="#" class="alert-link">endorse me for data science on LinkedIn</a>.
+</div> -->
+<section id="content" class="body">
+    <header>
+    <h1>
+      Reverse an array
+    </h1>
+<ol class="breadcrumb">
+    <li>
+        <time class="published" datetime="2018-11-07T09:00:00+01:00">
+            07 novembre 2018
+        </time>
+    </li>
+    <li>Python</li>
+    <li>Basics</li>
+</ol>
+</header>
+<div class='article_content'>
+<h2>Using lists</h2>
+<p>Python provides an iterator to reverse an array <a href="https://www.python.org/dev/peps/pep-0322/">reversed()</a></p>
+<p>you can use it this way :</p>
+<div class="highlight"><pre><span></span><span class="n">arr</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">range</span><span class="p">(</span><span class="mi">5</span><span class="p">))</span>
+<span class="k">print</span><span class="p">(</span><span class="n">f</span><span class="s1">&#39;Initial array : {arr}</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+
+
+<span class="n">reversed_arr</span> <span class="o">=</span> <span class="nb">list</span><span class="p">(</span><span class="nb">reversed</span><span class="p">(</span><span class="n">arr</span><span class="p">))</span>
+<span class="k">print</span><span class="p">(</span><span class="n">f</span><span class="s1">&#39;Reversed array : {reversed_arr}&#39;</span><span class="p">)</span>
+</pre></div>
+
+
+<div class="highlight"><pre><span></span>Initial array : [0, 1, 2, 3, 4]
+
+Reversed array : [4, 3, 2, 1, 0]
+</pre></div>
+
+
+<h2>Using numpy</h2>
+<p>If you are using numpy you can reverse an array by creating a view on the array:</p>
+<div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">numpy</span> <span class="kn">as</span> <span class="nn">np</span>
+<span class="n">arr</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">arange</span><span class="p">(</span><span class="mi">5</span><span class="p">)</span>
+<span class="k">print</span><span class="p">(</span><span class="n">f</span><span class="s1">&#39;Initial array : {arr}</span><span class="se">\n</span><span class="s1">&#39;</span><span class="p">)</span>
+
+
+<span class="n">reversed_arr</span> <span class="o">=</span> <span class="n">arr</span><span class="p">[::</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span>
+<span class="k">print</span><span class="p">(</span><span class="n">f</span><span class="s1">&#39;Reversed array : {reversed_arr}&#39;</span><span class="p">)</span>
+</pre></div>
+
+
+<div class="highlight"><pre><span></span>Initial array : [0 1 2 3 4]
+
+Reversed array : [4 3 2 1 0]
+</pre></div>
+</div>
+    <aside>
+    <div class="bug-reporting__panel">
+        <h3>Find an error or bug? Have a suggestion?</h3>
+        <p>Everything on this site is avaliable on GitHub. Head on over and <a href='https://github.com/redoules/redoules.github.io/issues/new'>submit an issue.</a> You can also message me directly by <a href='mailto:guillaume.redoules@gadz.org'>email</a>.</p>
+    </div>
+    </aside>
+</section>
+
+    </div>
+    <!-- start of footer section -->
+    <footer class="footer">
+        <div class="container">
+            <p class="text-muted">
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <br/>
+                Copyright &copy; Guillaume Redoulès,
+                    <time datetime="2018">2018</time>.
+                </center>
+            </p>
+        </div>
+    </footer>
+
+    <!-- This jQuery line finds any span that contains code highlighting classes and then selects the parent <pre> tag and adds a border. This is done as a workaround to visually distinguish the code inputs and outputs -->
+    <script>
+        $( ".hll, .n, .c, .err, .k, .o, .cm, .cp, .c1, .cs, .gd, .ge, .gr, .gh, .gi, .go, .gp, .gs, .gu, .gt, .kc, .kd, .kn, .kp, .kr, .kt, .m, .s, .na, .nb, .nc, .no, .nd, .ni, .ne, .nf, .nl, .nn, .nt, .nv, .ow, .w, .mf, .mh, .mi, .mo, .sb, .sc, .sd, .s2, .se, .sh, .si, .sx, .sr, .s1, .ss, .bp, .vc, .vg, .vi, .il" ).parent( "pre" ).css( "border", "1px solid #DEDEDE" );
+    </script>
+
+    
+    <!-- Load Google Analytics -->
+    <script>
+        /*
+        (function(i, s, o, g, r, a, m) {
+            i['GoogleAnalyticsObject'] = r;
+            i[r] = i[r] || function() {
+                (i[r].q = i[r].q || []).push(arguments)
+            }, i[r].l = 1 * new Date();
+            a = s.createElement(o),
+                m = s.getElementsByTagName(o)[0];
+            a.async = 1;
+            a.src = g;
+            m.parentNode.insertBefore(a, m)
+        })(window, document, 'script', '//www.google-analytics.com/analytics.js', 'ga');
+
+        ga('create', 'UA-66582-32', 'auto');
+        ga('send', 'pageview');
+        */
+    </script>
+    <!-- End of Google Analytics -->
+
+    <!-- Bootstrap core JavaScript
+      ================================================== -->
+    <!-- Placed at the end of the document so the pages load faster -->
+    <script src="../theme/js/bootstrap.min.js"></script>
+    <!-- IE10 viewport hack for Surface/desktop Windows 8 bug -->
+    <script src="../theme/js/ie10-viewport-bug-workaround.js"></script>
+
+
+</body>
+
+</html>
\ No newline at end of file
diff --git a/python/stock_pandas.html b/python/stock_pandas.html
index dfd442b..b1ccb2d 100644
--- a/python/stock_pandas.html
+++ b/python/stock_pandas.html
@@ -381,7 +381,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/python/updating_all_python_package_with_anaconda.html b/python/updating_all_python_package_with_anaconda.html
index b0c9d37..ec8a466 100644
--- a/python/updating_all_python_package_with_anaconda.html
+++ b/python/updating_all_python_package_with_anaconda.html
@@ -151,7 +151,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/search.html b/search.html
index 7261c93..a365d45 100644
--- a/search.html
+++ b/search.html
@@ -112,7 +112,7 @@
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/sitemap.xml b/sitemap.xml
index 7218d9d..f488408 100644
--- a/sitemap.xml
+++ b/sitemap.xml
@@ -5,7 +5,7 @@ xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
 
 <url>
 <loc>redoules.github.io/</loc>
-<lastmod>2018-10-30T11:37:58-00:00</lastmod>
+<lastmod>2018-11-07T21:11:20-00:00</lastmod>
 <changefreq>daily</changefreq>
 <priority>0.5</priority>
 </url>
@@ -17,6 +17,27 @@ xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
 <priority>0.5</priority>
 </url>
 
+<url>
+<loc>redoules.github.io/linux/simple_bash_function.html</loc>
+<lastmod>2018-11-07T09:37:00+01:00</lastmod>
+<changefreq>monthly</changefreq>
+<priority>0.5</priority>
+</url>
+
+<url>
+<loc>redoules.github.io/mathematics/Number_edges_Complete_graph.html</loc>
+<lastmod>2018-11-07T09:23:00+01:00</lastmod>
+<changefreq>monthly</changefreq>
+<priority>0.5</priority>
+</url>
+
+<url>
+<loc>redoules.github.io/python/reverse.html</loc>
+<lastmod>2018-11-07T09:00:00+01:00</lastmod>
+<changefreq>monthly</changefreq>
+<priority>0.5</priority>
+</url>
+
 <url>
 <loc>redoules.github.io/python/design_own_libs.html</loc>
 <lastmod>2018-10-29T20:31:00+01:00</lastmod>
diff --git a/sql/Sorting_results.html b/sql/Sorting_results.html
index 34c58d7..914b36f 100644
--- a/sql/Sorting_results.html
+++ b/sql/Sorting_results.html
@@ -344,7 +344,7 @@ Done.
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/sql/WHERE_SQL_keywords.html b/sql/WHERE_SQL_keywords.html
index 05fe931..46cfb3c 100644
--- a/sql/WHERE_SQL_keywords.html
+++ b/sql/WHERE_SQL_keywords.html
@@ -631,7 +631,7 @@ Done.
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/sql/display_table.html b/sql/display_table.html
index ba114d2..b23f742 100644
--- a/sql/display_table.html
+++ b/sql/display_table.html
@@ -216,7 +216,7 @@ Done.
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/sql/display_table_filter.html b/sql/display_table_filter.html
index 237f7d6..bd84bc7 100644
--- a/sql/display_table_filter.html
+++ b/sql/display_table_filter.html
@@ -346,7 +346,7 @@ Done.
     <footer class="footer">
         <div class="container">
             <p class="text-muted">
-                <center>This project contains 39 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
+                <center>This project contains 42 pages and is available on <a href="https://github.com/redoules/redoules.github.io">GitHub</a>.
                 <br/>
                 Copyright &copy; Guillaume Redoulès,
                     <time datetime="2018">2018</time>.
diff --git a/tipuesearch_content.json b/tipuesearch_content.json
index 84ed09b..3f627df 100644
--- a/tipuesearch_content.json
+++ b/tipuesearch_content.json
@@ -1 +1 @@
-{"pages":[{"title":"About Guillaume Redoulès","text":"I am a data scientist and a mechanical engineer working on numerical methods for stress computations in the field of rocket propulsion. Prior to that, I've got a MSc in Computational Fluid Dynamics and aerodynamics from Imperial College London. Email: guillaume.redoules@gadz.org Linkedin: Guillaume Redoulès Curriculum Vitae Experience Thermomecanical method and tools engineer , Ariane Group , 2015 - Present In charge of tools and methods related to thermomecanical computations. Focal point for machine learning. Education MSc Advanced Computational Methods for Aeronautics, Flow Management and Fluid-Structure Interaction , Imperial College London, London. 2013 Dissertation: \"Estimator design for fluid flows\" Fields: Aeronautics, aerodynamics, computational fluid dynamics, numerical methods Arts et Métiers Paristech , France, 2011 Generalist engineering degree Fields: Mechanics, electrical engineering, casting, machining, project management, finance, IT, etc.","tags":"pages","url":"redoules.github.io/pages/about.html","loc":"redoules.github.io/pages/about.html"},{"title":"Advice for designing your own libraries","text":"Advice for designing your own libraries When designing your own library make sure to think of the following things. I will add new paragraphs to this article as I dicover new good practices. Use standard python objects Try to use standard python objects as much as possible. That way, your library becomes compatible with all the other python libaries. For instance, when I created SAMpy : a library for reading and writing SAMCEF results, it returned dictonnaries, lists and pandas dataframes. Hence the results extracted from SAMCEF where compatible with all the scientific stack of python. Limit the number of functionnalities Following the same logic as before, the objects should do only one thing but do it well. Indeed, having a simple interface will reduce the complexity of your code and make it easier to use your library. Again, with SAMpy, I decided to strictly limit the functionnalities to reading and writing SAMCEF files. Define an exception class for your library You should define your own exceptions in order to make it easier for your users to debug their code thanks to clearer messages that convey more meaning. That way, the user will know if the error comes from your library or something else. Bonus if you group similar exceptions in a hierachy of inerited Exception classes. Example : let's create a Exception related to the age of a person : def check_age ( age ): if age < 0 and age > 130 : raise ValueError If the user inputed an invalid age, the ValueError exception would be thrown. That's fine but imagine you wan't to provide more feedback to your users that don't know the internal of your library. Let's now create a selfexplanatory Exception class AgeInvalidError ( ValueError ): pass def check_age ( age ): if age < 0 and age > 130 : raise AgeInvalidError ( age ) You can also add some helpful text to guide your users along the way: class AgeInvalidError ( ValueError ): print ( \"Age invalid, must be between 0 and 130\" ) pass def check_age ( age ): if age < 0 and age > 130 : raise AgeInvalidError ( age ) If you want to group all the logically linked exceptions, you can create a base class and inherit from it : class BaseAgeInvalidError ( ValueError ): pass class TooYoungError ( BaseAgeInvalidError ): pass class TooOldError ( BaseAgeInvalidError ): pass def check_age ( age ): if age < 0 : raise TooYoungError ( age ) elif age > 130 : raise TooOldError ( age ) Structure your repository You should have a file structure in your repository. It will help other contributers especially future contributers. A nice directory structure for your project should look like this: README.md LICENSE setup.py requirements.txt ./MyPackage ./docs ./tests Some prefer to use reStructured Text, I personnaly prefer Markdown choosealicense.com will help you pick the license to use for your project. For package and distribution management, create a setup.py file a the root of the directory The list of dependencies required to test, build and generate the doc are listed in a pip requirement file placed a the root of the directory and named requirements.txt Put the documentation of your library in the docs directory. Put your tests in the tests directory. Since your tests will need to import your library, I recommend modifying the path to resolve your package property. In order to do so, you can create a context.py file located in the tests directory : import os import sys sys . path . insert ( 0 , os . path . abspath ( os . path . join ( os . path . dirname ( __file__ ), '..' ))) import MyPackage Then within your individual test files you can import your package like so : from .context import MyPackage Finally, your code will go into the MyPackage directory Test your code Once your library is in production, you have to guaranty some level of forward compatibility. Once your interface is defined, write some tests. In the future, when your code is modified, having those tests will make sure that the behaviour of your functions and objects won't be altered. Document your code Of course, you should have a documentation to go along with your library. Make sure to add a lot of commun examples as most users tend to learn from examples. I recommend writing your documentation using Sphinx.","tags":"Python","url":"redoules.github.io/python/design_own_libs.html","loc":"redoules.github.io/python/design_own_libs.html"},{"title":"Safely creating a folder if it doesn't exist","text":"Safely creating a folder if it doesn't exist When you are writing to files in python, if the file doesn't exist it will be created. However, if you are trying to write a file in a directory that doesn't exist, an exception will be returned FileNotFoundError : [ Errno 2 ] No such file or directory : \"directory\" This article will teach you how to make sure the target directory exists. If it doesn't, the function will create that directory. First, let's import os and make sure that the \"test_directory\" doesn't exist import os os . path . exists ( \". \\\\ test_directory\" ) False copy the ensure_dir function in your code. This function will handle the creation of the directory. Credit goes to Parand posted on StackOverflow def ensure_dir ( file_path ): directory = os . path . dirname ( file_path ) if not os . path . exists ( directory ): os . makedirs ( directory ) Let's now use the function and create a folder named \"test_directory\" ensure_dir ( \". \\\\ test_directory\" ) If we test for the existence of the directory, the exists function will now return True os . path . exists ( \". \\\\ test_directory\" ) True","tags":"Python","url":"redoules.github.io/python/ensure_dir.html","loc":"redoules.github.io/python/ensure_dir.html"},{"title":"List all files in a directory","text":"Listing all the files in a directory Let's start with the basics, the most staigthforward way to list all the files in a direcoty is to use a combinaison of the listdir function and isfile form os.path. You can use a list comprehension to store all the results in a list. mypath = \"./test_directory/\" from os import listdir from os.path import isfile , join [ f for f in listdir ( mypath ) if isfile ( join ( mypath , f ))] ['logfile.log', 'myfile.txt', 'super_music.mp3', 'textfile.txt'] Listing all the files of a certain type in a directory similarly, if you want to filter only a certain kind of file based on its extension you can use the endswith method. In the following example, we will filter all the \"txt\" files contained in the directory [ f for f in listdir ( mypath ) if f . endswith ( '.' + \"txt\" )] ['myfile.txt', 'textfile.txt'] Listing all the files matching a pattern in a directory The glob module finds all the pathnames matching a specified pattern according to the rules used by the Unix shell. You can use the *, ?, and character ranges expressed with [] wildcards import glob glob . glob ( \"*.txt\" ) ['myfile.txt'] Listing files recusively If you want to list all files recursively you can select all the sub-directories using the \"**\" wildcard import glob glob . glob ( mypath + '/**/*.txt' , recursive = True ) ['./test_directory\\\\myfile.txt', './test_directory\\\\textfile.txt', './test_directory\\\\subdir1\\\\file_hidden_in_a_sub_direcotry.txt'] Using a regular expression If you'd rather use a regular expression to select the files, the pathlib library provides the rglob function. from pathlib import Path list ( Path ( \"./test_directory/\" ) . rglob ( \"*.[tT][xX][tT]\" )) [WindowsPath('test_directory/myfile.txt'), WindowsPath('test_directory/textfile.txt'), WindowsPath('test_directory/subdir1/file_hidden_in_a_sub_direcotry.txt')] Using regular expressions you can for example select multiple types of files. In the following example, we list all the files that finish either with \"txt\" or with \"log\". list ( Path ( \"./test_directory/\" ) . rglob ( \"*.[tl][xo][tg]\" )) [WindowsPath('test_directory/logfile.log'), WindowsPath('test_directory/myfile.txt'), WindowsPath('test_directory/textfile.txt'), WindowsPath('test_directory/subdir1/file_hidden_in_a_sub_direcotry.txt')]","tags":"Python","url":"redoules.github.io/python/list_files_directory.html","loc":"redoules.github.io/python/list_files_directory.html"},{"title":"Using Dask on infiniband","text":"InfiniBand (abbreviated IB) is a computer-networking communications standard used in high-performance computing that features very high throughput and very low latency. It is used for data interconnect both among and within computers. InfiniBand is also used as either a direct or switched interconnect between servers and storage systems, as well as an interconnect between storage systems. (source Wikipedia). If you want to leverage this high speed network instead of the regular ethernet network, you have to specify to the scheduler that you want to used infiniband as your interface. Assuming that you Infiniband interface is ib0 , you would call the scheduler like this : dask-scheduler --interface ib0 --scheduler-file ./cluster.yaml you would have to call the worker using the same interface : dask-worker --interface ib0 --scheduler-file ./cluster.yaml","tags":"Python","url":"redoules.github.io/python/dask_infiniband.html","loc":"redoules.github.io/python/dask_infiniband.html"},{"title":"Clearing the current cell in the notebook","text":"In python, you can clear the output of a cell by importing the IPython.display module and using the clear_output function from IPython.display import clear_output print ( \"text to be cleared\" ) clear_output () As you can see, the text \"text to be cleared\" is not displayed because the function clear_output has been called afterward","tags":"Jupyter","url":"redoules.github.io/jupyter/clear_cell.html","loc":"redoules.github.io/jupyter/clear_cell.html"},{"title":"What's inside my .bashrc ?","text":"############ # Anaconda # ############ export PATH = \"/station/guillaume/anaconda3/bin: $PATH \" alias python = '/station/guillaume/anaconda3/bin/python' ######### # Alias # ######### alias ding = 'echo -e \"\\a\"' alias calc = 'python -ic \"from __future__ import division; from math import *\"' alias h = \"history|grep \" alias f = \"find . |grep \" alias p = \"ps aux |grep \" alias cdl = \"cd /data/guillaume\" alias cp = \"rsync -avz --progress\" alias grep = \"grep --color=auto\" alias ls = \"ls -hN --color=auto --group-directories-first\" alias ll = \"ls -hal\" alias sv = \"ssh compute_cluster\" alias ms = \"ls\" alias jl = \"jupyter lab\" alias lst = \"jupyter notebook list\" ########################## # bashrc personnalisation# ########################## force_color_prompt = yes export EDITOR = nano export BROWSER = \"firefox '%s' &\" if [ -n \" $force_color_prompt \" ] ; then if [ -x /usr/bin/tput ] && tput setaf 1 > & /dev/null ; then # We have color support; assume it's compliant with Ecma-48 # (ISO/IEC-6429). (Lack of such support is extremely rare, and such # a case would tend to support setf rather than setaf.) color_prompt = yes else color_prompt = fi fi if [ \" $color_prompt \" = yes ] ; then #\\h : hostname #\\u : user #\\w : current working directory #\\d : date #\\t : time yellow = 226 green = 83 pink = 198 blue = 34 PS1 = \"\\[\\033[38;5;22m\\]\\u\\[ $( tput sgr0 ) \\]\\[\\033[38;5;163m\\]@\\[ $( tput sgr0 ) \\]\\[\\033[38;5;22m\\]\\h\\[ $( tput sgr0 ) \\]\\[\\033[38;5;162m\\]:\\[ $( tput sgr0 ) \\]\\[\\033[38;5;172m\\]{\\[ $( tput sgr0 ) \\]\\[\\033[38;5;39m\\]\\w\\[ $( tput sgr0 ) \\]\\[\\033[38;5;172m\\]}\\[ $( tput sgr0 ) \\]\\[\\033[38;5;162m\\]>\\[ $( tput sgr0 ) \\]\"","tags":"Linux","url":"redoules.github.io/linux/bashrc.html","loc":"redoules.github.io/linux/bashrc.html"},{"title":"Efficient extraction of eigenvalues from a list of tensors","text":"When you manipulate FEM results you generally have either a: scalar field, vector field, * tensor field. With tensorial results, it is often useful to extract the eigenvalues in order to find the principal values. I have found that it is easier to store the components of the tensors in a 6 column pandas dataframe (because of the symmetric property of stress and strain tensors) import pandas as pd node = [ 1001 , 1002 , 1003 , 1004 ] #when dealing with FEM results you should remember at which element/node the result is computed (in the example, let's assume that we look at node from 1001 to 1004) tensor1 = [ 1 , 1 , 1 , 0 , 0 , 0 ] #eigen : 1 tensor2 = [ 4 , - 1 , 0 , 2 , 2 , 1 ] #eigen : 5.58443, -1.77931, -0.805118 tensor3 = [ 1 , 6 , 5 , 3 , 3 , 1 ] #eigen : 8.85036, 4.46542, -1.31577 tensor4 = [ 1 , 2 , 3 , 0 , 0 , 0 ] #eigen : 1, 2, 3 df = pd . DataFrame ([ tensor1 , tensor2 , tensor3 , tensor4 ], columns = [ \"XX\" , \"YY\" , \"ZZ\" , \"XY\" , \"XZ\" , \"YZ\" ]) df . index = node df .dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } XX YY ZZ XY XZ YZ 1001 1 1 1 0 0 0 1002 4 -1 0 2 2 1 1003 1 6 5 3 3 1 1004 1 2 3 0 0 0 If you want to extract the eigenvalues of a tensor with numpy you have to pass a n by n ndarray to the eigenvalue function. In order to avoid having to loop over each node, this oneliner is highly optimized and will help you invert a large number of tensors efficiently. The steps are basically, create a list of n by n values (here n=3) in the right order => reshape it to a list of tensors => pass it to the eigenvals function import numpy as np from numpy import linalg as LA eigenvals = LA . eigvals ( df [[ \"XX\" , \"XY\" , \"XZ\" , \"XY\" , \"YY\" , \"YZ\" , \"XZ\" , \"YZ\" , \"ZZ\" ]] . values . reshape ( len ( df ), 3 , 3 )) eigenvals array([[ 1. , 1. , 1. ], [ 5.58442834, -0.80511809, -1.77931025], [-1.31577211, 8.85035616, 4.46541595], [ 1. , 2. , 3. ]])","tags":"Python","url":"redoules.github.io/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html","loc":"redoules.github.io/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html"},{"title":"Optimized numpy random number generation on Intel CPU","text":"Python Intel distribution Make sure you have a python intel distribution. When you startup python you should see somethine like : Python 3.6.2 |Intel Corporation| (default, Aug 15 2017, 11:34:02) [MSC v.1900 64 bit (AMD64)] Type 'copyright', 'credits' or 'license' for more information IPython 6.1.0 -- An enhanced Interactive Python. Type '?' for help. If not, you can force the installation of the intel optimized python with : conda update --all conda config --add channels intel conda install numpy --channel intel --override-channels oh and by the way, make sure you a running an Intel CPU ;) Comparing numpy.random with numpy.random_intel Let's now test both the rand function with and without the Intel optimization import numpy as np from numpy import random , random_intel % timeit np . random . rand ( 10 ** 5 ) 1.06 ms ± 91.5 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each) % timeit np . random_intel . rand ( 10 ** 5 ) 225 µs ± 3.46 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)","tags":"Python","url":"redoules.github.io/python/Optimized_numpy_random_intel.html","loc":"redoules.github.io/python/Optimized_numpy_random_intel.html"},{"title":"How to check Linux process information?","text":"How to check Linux process information (CPU usage, memory, user information, etc.)? You need to use the ps command combined with the grep command. In the example, we want to check the information on the nginx process : ps aux | grep nginx It would return the output : root 9976 0.0 0.0 12272 108 ? S<s Aug12 0:00 nginx: master process /usr/bin/nginx -g pid /run/nginx.pid; daemon on; master_process on; http 16780 0.0 0.0 12384 684 ? S< Aug12 4:11 nginx: worker process http 16781 0.0 0.0 12556 708 ? S< Aug12 0:24 nginx: worker process http 16782 0.0 0.1 12292 744 ? S< Aug12 2:43 nginx: worker process http 16783 0.0 0.1 12276 872 ? S< Aug12 0:24 nginx: worker process admin 17612 0.0 0.1 5120 864 pts/4 S+ 11:22 0:00 grep --color=auto nginx The columns have the following order : USER;PID;%CPU;%MEM;VSZ;RSS;TTY;STAT;START;TIME;COMMAND USER = user owning the process PID = process ID of the process %CPU = It is the CPU time used divided by the time the process has been running. %MEM = ratio of the process's resident set size to the physical memory on the machine VSZ = virtual memory usage of entire process (in KiB) RSS = resident set size, the non-swapped physical memory that a task has used (in KiB) TTY = controlling tty (terminal) STAT = multi-character process state START = starting time or date of the process TIME = cumulative CPU time COMMAND = command with all its arguments Interactive display If you want an interactive display showing in real time the statistics of the running process you can use the top command. If htop is available on you system, use this instead.","tags":"Linux","url":"redoules.github.io/linux/linux_process_information.html","loc":"redoules.github.io/linux/linux_process_information.html"},{"title":"Check the size of a directory","text":"How do you Check the size of a directory in linux? The du command will come handy for this task. Let's say we want to know the size of the directory named recommandations , we would run the following command du -sh recommendations It would return the output : 9.9M recommendations","tags":"Linux","url":"redoules.github.io/linux/directory_size.html","loc":"redoules.github.io/linux/directory_size.html"},{"title":"Check for free disk space","text":"How do you check for free disk space on linux? The df command will come handy for this task. Run the command with the following arguments df -ah Will return in a human readable format for all drives a readout of all your filesystems Filesystem Size Used Avail Use% Mounted on /dev/md0 2.4G 1.3G 1.1G 54% / none 348M 4.0K 348M 1% /dev none 0 0 0 - /dev/pts none 0 0 0 - /proc none 0 0 0 - /sys /tmp 350M 1.3M 349M 1% /tmp /run 350M 3.2M 347M 1% /run /dev/shm 350M 12K 350M 1% /dev/shm /proc/bus/usb 0 0 0 - /proc/bus/usb securityfs 0 0 0 - /sys/kernel/security /dev/md3 1.8T 372G 1.5T 21% /volume2 /dev/vg1000/lv 1.8T 1.5T 340G 82% /volume1 /dev/sdq1 7.4G 3.8G 3.5G 52% /volumeUSB3/usbshare3-1 /dev/sdr 294G 146G 134G 53% /volumeUSB2/usbshare none 0 0 0 - /proc/fs/nfsd none 0 0 0 - /config The free disk space can be read in the Avail column","tags":"Linux","url":"redoules.github.io/linux/free_disk_space_linux.html","loc":"redoules.github.io/linux/free_disk_space_linux.html"},{"title":"Check your current ip address","text":"Check your current ip address Run the command ip addr show and that will give you every information available 4: eth0: <> mtu 1500 group default qlen 1 link/ether 5c:51:4f:41:7a:b1 inet 169.254.33.33/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::390a:f69e:1ba2:2121/64 scope global dynamic valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 group default qlen 1 link/ether 0a:00:27:00:00:03 inet 192.168.56.1/24 brd 192.168.56.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::84cd:374e:843f:f82f/64 scope global dynamic valid_lft forever preferred_lft forever 15: eth2: <> mtu 1500 group default qlen 1 link/ether 00:ff:d2:8a:19:c3 inet 169.254.40.62/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::f199:2ca0:7aff:283e/64 scope global dynamic valid_lft forever preferred_lft forever 1: lo: <LOOPBACK,UP> mtu 1500 group default qlen 1 link/loopback 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 ::1/128 scope global dynamic valid_lft forever preferred_lft forever 5: wifi0: <BROADCAST,MULTICAST,UP> mtu 1500 group default qlen 1 link/ieee802.11 5c:51:4f:41:7a:ad inet 192.168.1.1/24 brd 192.168.1.255 scope global dynamic valid_lft 42720sec preferred_lft 42720sec inet6 fe80::395f:3594:1dc2:57e3/64 scope global dynamic valid_lft forever preferred_lft forever 21: wifi1: <> mtu 1500 group default qlen 1 link/ieee802.11 5c:51:4f:41:7a:ae inet 169.254.12.77/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::58d5:630:cbbd:c4d/64 scope global dynamic valid_lft forever preferred_lft forever 12: eth3: <> mtu 1472 group default qlen 1 link/ether 00:00:00:00:00:00:00:e0:00:00:00:00:00:00:00:00 inet6 fe80::100:7f:fffe/64 scope global dynamic valid_lft forever preferred_lft forever 10: eth4: <BROADCAST,MULTICAST,UP> mtu 1500 group default qlen 1 link/ether 22:b7:57:52:5f:ff inet 192.168.42.106/24 brd 192.168.42.255 scope global dynamic valid_lft 6659sec preferred_lft 6659sec inet6 fe80::5110:eb6f:deb0:45c4/64 scope global dynamic valid_lft forever preferred_lft forever You can select only one interface ip addr show eth0 and only the relevant information will be displayed 4: eth0: <> mtu 1500 group default qlen 1 link/ether 5c:51:4f:41:7a:b1 inet 169.254.33.33/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::390a:f69e:1ba2:2121/64 scope global dynamic valid_lft forever preferred_lft forever Check your current ip address the old way The ifconfig command will return information regarding your network interfaces. Let's try it: ifconfig eth1 Link encap:Ethernet HWaddr 0a:00:27:00:00:03 inet adr:192.168.56.1 Bcast:192.168.56.255 Masque:255.255.255.0 adr inet6: fe80::84cd:374e:843f:f82f/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) eth4 Link encap:Ethernet HWaddr 22:b7:57:52:5f:ff inet adr:192.168.42.106 Bcast:192.168.42.255 Masque:255.255.255.0 adr inet6: fe80::5110:eb6f:deb0:45c4/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) lo Link encap:Boucle locale inet adr:127.0.0.1 Masque:255.0.0.0 adr inet6: ::1/128 Scope:Global UP LOOPBACK RUNNING MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) wifi0 Link encap:UNSPEC HWaddr 5C-51-4F-41-7A-AD-00-00-00-00-00-00-00-00-00-00 inet adr:192.168.1.1 Bcast:192.168.1.255 Masque:255.255.255.0 adr inet6: fe80::395f:3594:1dc2:57e3/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) On the left column you have the list of network adapters. The lo is the local loopback, it is an interface that points to the localhost. Interfaces starting with eth refer to wired connections over ethernet (or sometimes USB in the case of a phone acting as an access point over USB). Interfaces starting with wlan or wifi refer to wireless connections. On the right column you have some information corresponding to the interface such as the IPv4, the IPv6, the mask, some statistics about the interface and so on.","tags":"Linux","url":"redoules.github.io/linux/get_ip_linux.html","loc":"redoules.github.io/linux/get_ip_linux.html"},{"title":"Check the version of the kernel currently running","text":"Check the version of the kernel currently running The uname command will give you the version of the kernel. In order to get a more useful output, type uname -a This will return : the hostname os name kernel release/version architecture * etc. variations If you only want the kernel version you can type uname -v if you only want the kernel release you can type uname -r","tags":"Linux","url":"redoules.github.io/linux/version_kernel.html","loc":"redoules.github.io/linux/version_kernel.html"},{"title":"Running the notebook on a remote server","text":"Jupyter hub With JupyterHub you can create a multi-user Hub which spawns, manages, and proxies multiple instances of the single-user Jupyter notebook server. Project Jupyter created JupyterHub to support many users. The Hub can offer notebook servers to a class of students, a corporate data science workgroup, a scientific research project, or a high performance computing group. However, if you are the only one using the server and you just want a simple way to run the notebook on your server and access it through the web interface on a light client without having to install and configure the jupyter hub, you can do the following. Problem with jupyter notebook On your server, run the command jupyter-notebook you should get something like : [I 11:18:44.514 NotebookApp] Serving notebooks from local directory: /volume2/homes/admin [I 11:18:44.515 NotebookApp] 0 active kernels [I 11:18:44.516 NotebookApp] The Jupyter Notebook is running at: http://localhost:8888/?token=357587e3269b0f20f2b7e1918492890ae7573ac7ef1d2023 [I 11:18:44.516 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). [W 11:18:44.519 NotebookApp] No web browser found: could not locate runnable browser. [C 11:18:44.520 NotebookApp] Copy/paste this URL into your browser when you connect for the first time, to login with a token: http://localhost:8888/?token=357587e3269b0f20f2b7e1918492890ae7573ac7ef1d2023 and if you try to connect to your server ip (in my example : http://192.168.1.2:8888/?token=357587e3269b0f20f2b7e1918492890ae7573ac7ef1d2023) you will get an \"ERR_CONNECTION_REFUSED\" error. This is because, by default, Jupyter Notebook only accepts connections from localhost. Allowing connexions from other sources From any IP The simplest way to avoid the connection error is to allow the notebook to accept connections from any ip jupyter-notebook --ip = * you will get something like [W 11:26:45.285 NotebookApp] WARNING: The notebook server is listening on all IP addresses and not using encryption. This is not recommended. [I 11:26:45.342 NotebookApp] Serving notebooks from local directory: /volume2/homes/admin [I 11:26:45.342 NotebookApp] 0 active kernels [I 11:26:45.343 NotebookApp] The Jupyter Notebook is running at: http://[all ip addresses on your system]:8888/?token=52af33d628881824968b4031967e8541a27cc28b1720c199 [I 11:26:45.343 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). [W 11:26:45.346 NotebookApp] No web browser found: could not locate runnable browser. [C 11:26:45.347 NotebookApp] Copy/paste this URL into your browser when you connect for the first time, to login with a token: http://localhost:8888/?token=52af33d628881824968b4031967e8541a27cc28b1720c199 and if you connect form a remote client (192.168.1.1 in my example), the following line will be added to the output : [I 11:26:54.798 NotebookApp] 302 GET /?token=52af33d628881824968b4031967e8541a27cc28b1720c199 (192.168.1.1) 111.17ms note that you should only do that if you are the only one using the server because the connection is not encypted. From a specific IP You can also, explicitly specify the ip of the client jupyter-notebook --ip = 192 .168.1.1 [I 11:44:58.104 NotebookApp] JupyterLab extension loaded from C:\\Users\\Guillaume\\Miniconda3\\lib\\site-packages\\jupyterlab [I 11:44:58.104 NotebookApp] JupyterLab application directory is C:\\Users\\Guillaume\\Miniconda3\\share\\jupyter\\lab [I 11:44:58.244 NotebookApp] Serving notebooks from local directory: C:\\Users\\Guillaume [I 11:44:58.245 NotebookApp] 0 active kernels [I 11:44:58.245 NotebookApp] The Jupyter Notebook is running at: http://192.168.1.1:8888/?token=503576dd8fa87d1f2c416df307e9b900e520b4942e317b32 [I 11:44:58.245 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). [C 11:44:58.258 NotebookApp] Copy/paste this URL into your browser when you connect for the first time, to login with a token: http://192.168.1.1:8888/?token=503576dd8fa87d1f2c416df307e9b900e520b4942e317b32 [I 11:44:59.083 NotebookApp] Accepting one-time-token-authenticated connection from 192.168.1.1","tags":"Jupyter","url":"redoules.github.io/jupyter/remote_run_notebook.html","loc":"redoules.github.io/jupyter/remote_run_notebook.html"},{"title":"Running multiple calls to a function in parallel with Dask","text":"Dask.distributed is a lightweight library for distributed computing in Python. It allows to create a compute graph. Dask distributed is architectured around 3 parts : the dask-scheduler the dask-worker(s) the dask client Dask architecture The Dask scheduler is a centrally managed, distributed, dynamic task scheduler. It recieves tasks from a/multiple client(s) and spread them across one or multiple dask-worker(s). Dask-scheduler is an event based asynchronous dynamic scheduler, meaning that mutliple clients can submit a list of task to be executed on multiple workers. Internally, the task are represented as a directed acyclic graph. Both new clients and new workers can be connected or disconnected during the execution of the task graph. Tasks can be submited with the function client . submit ( function , * args , ** kwargs ) or by using objects from the dask library such as dask.dataframe, dask.bag or dask.array Setup In this example, we will use a distributed scheduler on a single machine with multiple workers and a single client. We will use the client to submit some tasks to the scheduler. The scheduler will then dispatch those tasks to the workers. The process can be monitored in real time through a web application. For this example, all the computations will be run on a local computer. However dask can scale to a large HPC cluster. First we have to launch the dask-scheduler; from the command line, input dask-scheduler Next, you can load the web dashboard. In order to do so, the scheduler returns the number of the port you have to connect to in the line starting with \"bokeh at :\". The default port is 8787. Since we are running all the programs on the same computer, we just have to login to http://127.0.0.1:8787/status Finally, we have to launch the dask-worker(s). If you want to run the worker(s) on the same computer as the scheduler the type : dask-worker 127 .0.0.1:8786 otherwise, make sure you are inputing the ip address of the computer hosting the dask-scheduler. You can launch as many workers as you want. In this example, we will run 3 workers on the local machine. Use the dask workers within your python code We will now see how to submit multiple calls to a fucntion in parallel on the dask-workers. Import the required libraries and define the function to be executed. import numpy as np import pandas as pd from distributed import Client #function used to do parallel computing on def compute_pi_MonteCarlo ( Nb_Data ): \"\"\" computes the value of pi using the monte carlo method \"\"\" Radius = 1 Nb_Data = int ( round ( Nb_Data )) x = np . random . uniform ( - Radius , Radius , Nb_Data ) y = np . random . uniform ( - Radius , Radius , Nb_Data ) pi_mc = 4 * np . sum ( np . power ( x , 2 ) + np . power ( y , 2 ) < Radius ** 2 ) / Nb_Data err = 100 * np . abs ( pi_mc - np . pi ) / np . pi return [ Nb_Data , pi_mc , err ] In order to connect to the scheduler, we create a client. client = Client ( '127.0.0.1:8786' ) client Client Scheduler: tcp://127.0.0.1:8786 Dashboard: http://127.0.0.1:8787/status Cluster Workers: 3 Cores: 12 Memory: 25.48 GB We submit tasks using the submit method data = [ client . submit ( compute_pi_MonteCarlo , Nb_Data ) for Nb_Data in np . logspace ( 3 , 7 , num = 1200 , dtype = int )] If you look at http://127.0.0.1:8787/status you will see the tasks beeing completed. Once competed, gather the data: data = client . gather ( data ) df = pd . DataFrame ( data ) df . columns = [ \"number of points for MonteCarlo\" , \"value of pi\" , \"error (%)\" ] df . tail () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } number of points for MonteCarlo value of pi error (%) 1195 9697405 3.141296 0.009454 1196 9772184 3.141058 0.017008 1197 9847540 3.141616 0.000739 1198 9923477 3.141009 0.018574 1199 10000000 3.141032 0.017833 There, we have completed a simple example on how to use dask to run multiple functions in parallel. Full source code: import numpy as np import pandas as pd from distributed import Client #function used to do parallel computing on def compute_pi_MonteCarlo ( Nb_Data ): \"\"\" computes the value of pi using the monte carlo method \"\"\" Radius = 1 Nb_Data = int ( round ( Nb_Data )) x = np . random . uniform ( - Radius , Radius , Nb_Data ) y = np . random . uniform ( - Radius , Radius , Nb_Data ) pi_mc = 4 * np . sum ( np . power ( x , 2 ) + np . power ( y , 2 ) < Radius ** 2 ) / Nb_Data err = 100 * np . abs ( pi_mc - np . pi ) / np . pi return [ Nb_Data , pi_mc , err ] #connect to the scheduler client = Client ( '127.0.0.1:8786' ) #submit tasks data = [ client . submit ( compute_pi_MonteCarlo , Nb_Data ) for Nb_Data in np . logspace ( 3 , 7 , num = 1200 , dtype = int )] #gather the results data = client . gather ( data ) df = pd . DataFrame ( data ) df . columns = [ \"number of points for MonteCarlo\" , \"value of pi\" , \"error (%)\" ] df . tail () A word on the environement variables On Windows, to make sure that you can run dask-scheduler and dask-worker from the command line, you have to add the location of the executable to your path. On linux, you can append the location of the dask-worker and scheduler to the path variable with the command export PATH = $PATH :/path/to/dask","tags":"Python","url":"redoules.github.io/python/dask_distributed_parallelism.html","loc":"redoules.github.io/python/dask_distributed_parallelism.html"},{"title":"Plotting data using log axis","text":"Plotting in log axis with matplotlib import matplotlib.pyplot as plt % matplotlib inline import numpy as np x = np . linspace ( 0.1 , 20 ) y = 20 * np . exp ( - x / 10.0 ) Plotting using the standard function then specifying the axis scale One of the easiest way to plot in a log plot is to specify the plot normally and then specify which axis is to be plotted with a log scale. This can be specified by the function set_xscale or set_yscale # Normal plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . grid () plt . show () # Log x axis plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . set_xscale ( 'log' ) ax . grid () plt . show () # Log x axis plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . set_yscale ( 'log' ) ax . grid () plt . show () # Log x axis plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . set_xscale ( 'log' ) ax . set_yscale ( 'log' ) ax . grid () plt . show () Plotting using the matplotlib defined function Matplotlib has the function : semilogx, semilogy and loglog that can help you avoid having to specify the axis scale. # Plot using semilogx fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . semilogx ( x , y ) ax . grid () plt . show () # Plot using semilogy fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . semilogy ( x , y ) ax . grid () plt . show () # Plot using loglog fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . loglog ( x , y ) ax . grid () plt . show ()","tags":"Python","url":"redoules.github.io/python/logplot.html","loc":"redoules.github.io/python/logplot.html"},{"title":"Downloading a static webpage with python","text":"If you are using python legacy (aka python 2) first of all, stop ! Furthermore, this method won't work in python legacy # Import modules from urllib.request import urlopen The webpage source code can be downloaded with the command urlopen url = \"http://example.com/\" #create a HTTP request in order to read the page page = urlopen ( url ) . read () The source code will be stored in the variable page as a string print ( page ) b '&lt;!doctype html&gt;\\n&lt;html&gt;\\n&lt;head&gt;\\n &lt;title&gt;Example Domain&lt;/title&gt;\\n\\n &lt;meta charset=\"utf-8\" /&gt;\\n &lt;meta http-equiv=\"Content-type\" content=\"text/html; charset=utf-8\" /&gt;\\n &lt;meta name=\"viewport\" content=\"width=device-width, initial-scale=1\" /&gt;\\n &lt;style type=\"text/css\"&gt;\\n body {\\n background-color: #f0f0f2;\\n margin: 0;\\n padding: 0;\\n font-family: \"Open Sans\", \"Helvetica Neue\", Helvetica, Arial, sans-serif;\\n \\n }\\n div {\\n width: 600px;\\n margin: 5em auto;\\n padding: 50px;\\n background-color: #fff;\\n border-radius: 1em;\\n }\\n a:link, a:visited {\\n color: #38488f;\\n text-decoration: none;\\n }\\n @media (max-width: 700px) {\\n body {\\n background-color: #fff;\\n }\\n div {\\n width: auto;\\n margin: 0 auto;\\n border-radius: 0;\\n padding: 1em;\\n }\\n }\\n &lt;/style&gt; \\n&lt;/head&gt;\\n\\n&lt;body&gt;\\n&lt;div&gt;\\n &lt;h1&gt;Example Domain&lt;/h1&gt;\\n &lt;p&gt;This domain is established to be used for illustrative examples in documents. You may use this\\n domain in examples without prior coordination or asking for permission.&lt;/p&gt;\\n &lt;p&gt;&lt;a href=\"http://www.iana.org/domains/example\"&gt;More information...&lt;/a&gt;&lt;/p&gt;\\n&lt;/div&gt;\\n&lt;/body&gt;\\n&lt;/html&gt;\\n' Additionally, you can beautifulsoup in order to make it easier to work with html from bs4 import BeautifulSoup soup = BeautifulSoup ( page , 'lxml' ) soup . prettify () print ( soup ) & lt ;! DOCTYPE html & gt ; & lt ; html & gt ; & lt ; head & gt ; & lt ; title & gt ; Example Domain & lt ;/ title & gt ; & lt ; meta charset = \"utf-8\" /& gt ; & lt ; meta content = \"text/html; charset=utf-8\" http-equiv = \"Content-type\" /& gt ; & lt ; meta content = \"width=device-width, initial-scale=1\" name = \"viewport\" /& gt ; & lt ; style type = \"text/css\" & gt ; body { background-color : #f0f0f2 ; margin : 0 ; padding : 0 ; font-family : \"Open Sans\" , \"Helvetica Neue\" , Helvetica , Arial , sans-serif ; } div { width : 600 px ; margin : 5 em auto ; padding : 50 px ; background-color : #fff ; border-radius : 1 em ; } a : link , a : visited { color : #38488f ; text-decoration : none ; } @ media ( max-width : 700px ) { body { background-color : #fff ; } div { width : auto ; margin : 0 auto ; border-radius : 0 ; padding : 1 em ; } } & lt ;/ style & gt ; & lt ;/ head & gt ; & lt ; body & gt ; & lt ; div & gt ; & lt ; h1 & gt ; Example Domain & lt ;/ h1 & gt ; & lt ; p & gt ; This domain is established to be used for illustrative examples in documents . You may use this domain in examples without prior coordination or asking for permission .& lt ;/ p & gt ; & lt ; p & gt ;& lt ; a href = \"http://www.iana.org/domains/example\" & gt ; More information ...& lt ;/ a & gt ;& lt ;/ p & gt ; & lt ;/ div & gt ; & lt ;/ body & gt ; & lt ;/ html & gt ;","tags":"Python","url":"redoules.github.io/python/download_page.html","loc":"redoules.github.io/python/download_page.html"},{"title":"Getting stock market data","text":"Start by importing the packages. We will need pandas and the pandas_datareader. # Import modules import pandas as pd from pandas_datareader import data Datareader allows you to import data from the internet. I have found that Quandl and robinhood works the best as a source for stockmarket data. Note that if you want an other type of data (e.g. GDP, inflation, etc.) other sources exist. #import stock from robinhood aapl_robinhood = data . DataReader ( 'AAPL' , 'robinhood' , '1980-01-01' ) aapl_robinhood . head () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } close_price high_price interpolated low_price open_price session volume symbol begins_at AAPL 2017-08-04 153.996200 154.990700 False 153.306900 153.681100 reg 20559852 2017-08-07 156.379100 156.487400 False 154.272000 154.655900 reg 21870321 2017-08-08 157.629700 159.352900 False 155.847400 156.172300 reg 36205896 2017-08-09 158.594700 158.801500 False 156.674500 156.822200 reg 26131530 2017-08-10 153.543100 158.169600 False 152.861000 158.070700 reg 40804273 #import stock from quandl aapl_quandl = data . DataReader ( 'AAPL' , 'quandl' , '1980-01-01' ) aapl_quandl . head () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } Open High Low Close Volume ExDividend SplitRatio AdjOpen AdjHigh AdjLow AdjClose AdjVolume Date 2018-03-27 173.68 175.15 166.92 168.340 38962839.0 0.0 1.0 173.68 175.15 166.92 168.340 38962839.0 2018-03-26 168.07 173.10 166.44 172.770 36272617.0 0.0 1.0 168.07 173.10 166.44 172.770 36272617.0 2018-03-23 168.39 169.92 164.94 164.940 40248954.0 0.0 1.0 168.39 169.92 164.94 164.940 40248954.0 2018-03-22 170.00 172.68 168.60 168.845 41051076.0 0.0 1.0 170.00 172.68 168.60 168.845 41051076.0 2018-03-21 175.04 175.09 171.26 171.270 35247358.0 0.0 1.0 175.04 175.09 171.26 171.270 35247358.0","tags":"Python","url":"redoules.github.io/python/stock_pandas.html","loc":"redoules.github.io/python/stock_pandas.html"},{"title":"Moving average with pandas","text":"# Import modules import pandas as pd from pandas_datareader import data , wb #import packages from pandas_datareader import data aapl = data . DataReader ( 'AAPL' , 'quandl' , '1980-01-01' ) aapl . head () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } Open High Low Close Volume ExDividend SplitRatio AdjOpen AdjHigh AdjLow AdjClose AdjVolume Date 2018-03-27 173.68 175.15 166.92 168.340 38962839.0 0.0 1.0 173.68 175.15 166.92 168.340 38962839.0 2018-03-26 168.07 173.10 166.44 172.770 36272617.0 0.0 1.0 168.07 173.10 166.44 172.770 36272617.0 2018-03-23 168.39 169.92 164.94 164.940 40248954.0 0.0 1.0 168.39 169.92 164.94 164.940 40248954.0 2018-03-22 170.00 172.68 168.60 168.845 41051076.0 0.0 1.0 170.00 172.68 168.60 168.845 41051076.0 2018-03-21 175.04 175.09 171.26 171.270 35247358.0 0.0 1.0 175.04 175.09 171.26 171.270 35247358.0 In order to computer the moving average, we will use the rolling function. #120 days moving average moving_averages = aapl [[ \"Open\" , \"High\" , \"Low\" , \"Close\" , \"Volume\" ]] . rolling ( window = 120 ) . mean () moving_averages . tail () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } Open High Low Close Volume Date 1980-12-18 28.457667 28.551917 28.385000 28.385000 139495.000000 1980-12-17 28.410750 28.502917 28.338083 28.338083 141772.500000 1980-12-16 28.362833 28.453917 28.289167 28.289167 141256.666667 1980-12-15 28.335750 28.426833 28.262083 28.262083 144321.666667 1980-12-12 28.310750 28.402833 28.238167 28.238167 159625.000000 % matplotlib inline import matplotlib.pyplot as plt plt . plot ( aapl . index , aapl . Open , label = 'Open price' ) plt . plot ( moving_averages . index , moving_averages . Open , label = \"120 MA Open price\" ) plt . legend () plt . show ()","tags":"Python","url":"redoules.github.io/python/Moving_average_pandas.html","loc":"redoules.github.io/python/Moving_average_pandas.html"},{"title":"Keywords to use with WHERE","text":"Keywords to use with WHERE #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' Assignment operator The assignment operator is =. % sql SELECT * FROM tutyfrutty WHERE color = \"red\" * sqlite:///mydatabase.db Done. index fruit color kcal 2 Apple red 52 7 Cranberry red 308 Comparison operators Comparison operation can be done in a SQL querry. They are the following : Equality : = Greater than : > greater than or equal to : >= less than : < less than or equal to : <= not equal to : <>, != not greater than : !> not less than : !< % sql SELECT * FROM tutyfrutty WHERE kcal = 47 * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 % sql SELECT * FROM tutyfrutty WHERE kcal > 47 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE kcal >= 47 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE kcal < 47 * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 5 plum purple 28 % sql SELECT * FROM tutyfrutty WHERE kcal <= 47 * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 3 lemon yellow 15 4 lime green 30 5 plum purple 28 % sql SELECT * FROM tutyfrutty WHERE kcal <> 47 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 7 Cranberry red 308 Logical operators Logical operators test a condition and return a boolean. The logicial operators in SQL are : ALL : true if all the condtions are true AND : true is both conditions are true ANY : true if any one of the conditions are true BETWEEN : true if the operand in withing a range of values EXISTS : true if the subquery contains any rows IN : true if the condition is present in a row LIKE : true if a pattern is matched NOT : True if the operand is false, false otherwise OR : True is either condition is true SOME : true is any of the conditions is true % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" AND kcal < 100 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 3 lemon yellow 15 % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR kcal > 300 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 3 lemon yellow 15 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE fruit LIKE 'l%' * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 % sql SELECT * FROM tutyfrutty WHERE NOT color = \"yellow\" * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 2 Apple red 52 4 lime green 30 5 plum purple 28 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE kcal BETWEEN 40 AND 100 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 Bitwise operators Some bitwise operators exist in SQL. They will not be demonstrated here. They are the following : AND : & OR : | XOR : &#94; NOT : ~","tags":"SQL","url":"redoules.github.io/sql/WHERE_SQL_keywords.html","loc":"redoules.github.io/sql/WHERE_SQL_keywords.html"},{"title":"Sorting results","text":"Sorting results in SQL Sorting results can be achieved by using a modifier command at the end of the SQL querry #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' The results can be sorted with the command ORDER BY SELECT column-list FROM table_name [WHERE condition] [ORDER BY column1, column2, .. columnN] [ASC | DESC] Let's show an example where we extract the fruits that are either yellow or red % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 7 Cranberry red 308 Ascending sort % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" ORDER BY kcal ASC * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 2 Apple red 52 0 Banana yellow 89 7 Cranberry red 308 descending sort % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" ORDER BY kcal DESC * sqlite:///mydatabase.db Done. index fruit color kcal 7 Cranberry red 308 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 Sort by multiple columns You can sort by more than one column. Just specify multiple columns in the ORDER BY keyword. In the example, we will sort alphabetically on the color column first and sort alphabetically on the fruit column % sql SELECT * FROM tutyfrutty ORDER BY color , fruit ASC * sqlite:///mydatabase.db Done. index fruit color kcal 4 lime green 30 1 Orange orange 47 5 plum purple 28 2 Apple red 52 7 Cranberry red 308 0 Banana yellow 89 3 lemon yellow 15","tags":"SQL","url":"redoules.github.io/sql/Sorting_results.html","loc":"redoules.github.io/sql/Sorting_results.html"},{"title":"Filter content of a TABLE","text":"Filter content of a TABLE in SQL In this example, we will display the content of a table but we will filter out the results. Since we are working in the notebook, we will load the sql extension in order to manipulate the database. The database mydatabase.db is a SQLite database already created before the example. #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' Filter content matching exactly a condition We want to extract all the entries in a dataframe that match a certain condition, in order to do so, we will use the following command : SELECT * FROM TABLE WHERE column=\"condition\" In our example, we will filter all the entries in the tutyfrutty table whose color is yellow % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 3 lemon yellow 15 Complex conditions You can build more complex conditions by using the keywords OR and AND In the following example, we will filter all entries that are either yellow or red % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 7 Cranberry red 308 Note : when combining multiple conditions with AND and OR, be careful to use parentesis where needed Conditions matching a pattern You can also use the LIKE keyword in order to find all entries that match a certain pattern. In our example, we want to find all fruits begining with a \"l\". In order to do so, we will use the LIKE keyword and the wildcard \"%\" meaning any string % sql SELECT * FROM tutyfrutty WHERE fruit LIKE \"l%\" * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 Numerical conditions When we are working with numerical data, we can use the GREATER THAN > and SMALLER THAN < operators % sql SELECT * FROM tutyfrutty WHERE kcal < 47 * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 5 plum purple 28 If we want the condition to be inclusive we can use the operator <= (alternatively >=) % sql SELECT * FROM tutyfrutty WHERE kcal <= 47 * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 3 lemon yellow 15 4 lime green 30 5 plum purple 28","tags":"SQL","url":"redoules.github.io/sql/display_table_filter.html","loc":"redoules.github.io/sql/display_table_filter.html"},{"title":"Displaying the content of a TABLE","text":"Displaying the content of a TABLE in SQL In this very simple example we will see how to display the content of a table. Since we are working in the notebook, we will load the sql extension in order to manipulate the database. The database mydatabase.db is a SQLite database already created before the example. #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' In order to extract all the values from a table, we will use the following command : SELECT * FROM TABLE In our example, we want to display the data contained in the table named tutyfrutty % sql SELECT * FROM tutyfrutty * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 7 Cranberry red 308","tags":"SQL","url":"redoules.github.io/sql/display_table.html","loc":"redoules.github.io/sql/display_table.html"},{"title":"Opening a file with python","text":"This short article show you how to open a file using python. We will use the with keyword in order to avoid having to close the file. There is no need to import anything in order to open a file. All the function related to file manipulation are part of the python standard library In order to open a file, we will use the function open. This function takes two arguments : the path of the file the mode you want to open the file The mode can be : 'r' : read 'w' : write 'a' : append (writes at the end of the file) 'b' : binary mode 'x' : exclusive creation 't' : text mode (by default) Note that if the file does not exit it will be created if you use the following options \"w\", \"a\", \"x\". If you try to open a non existing file in read mode 'r', a FileNotFoundError will be returned. It is possible to combine multiple options together. For instance, you can open a file in binary mode for writing using the 'wb' option. Python distinguishes between binary and text I/O. Files opened in binary mode return contents as bytes objects without any decoding. In text mode , the contents of the file are returned as str, the bytes having been first decoded using a platform-dependent encoding or using the specified encoding if given. Writing to a file Let's first open (create) a text file a write a string to it. filepath = \". \\\\ myfile.txt\" with open ( filepath , 'w' ) as f : f . write ( \"Hello world !\" ) Reading a file we can now see how to read the content of a file. To do so, we will use the 'r' option with open ( filepath , \"r\" ) as f : content = f . read () print ( content ) Hello world ! A word on the with keyword In python the with keyword is used when working with unmanaged resources (like file streams). The python documentation tells us that : The with statement clarifies code that previously would use try...finally blocks to ensure that clean-up code is executed. In this section, I'll discuss the statement as it will commonly be used. In the next section, I'll examine the implementation details and show how to write objects for use with this statement. The with statement is a control-flow structure whose basic structure is: with expression [ as variable ]: with - block The expression is evaluated, and it should result in an object that supports the context management protocol (that is, has enter () and exit () methods).","tags":"Python","url":"redoules.github.io/python/Opening_file.html","loc":"redoules.github.io/python/Opening_file.html"},{"title":"Opening a SQLite database with python","text":"This short article show you how to connect to a SQLite database using python. We will use the with keyword in order to avoid having to close the database. In order to connect to the database, we will have to import sqlite3 import sqlite3 from sqlite3 import Error In python the with keyword is used when working with unmanaged resources (like file streams). The python documentation tells us that : The with statement clarifies code that previously would use try...finally blocks to ensure that clean-up code is executed. In this section, I'll discuss the statement as it will commonly be used. In the next section, I'll examine the implementation details and show how to write objects for use with this statement. The with statement is a control-flow structure whose basic structure is: with expression [ as variable ]: with - block The expression is evaluated, and it should result in an object that supports the context management protocol (that is, has enter () and exit () methods). db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : print ( \"Connected to the database\" ) #your code here except Error as e : print ( e ) Connected to the database","tags":"Python","url":"redoules.github.io/python/Opening_SQLite_database.html","loc":"redoules.github.io/python/Opening_SQLite_database.html"},{"title":"Reading data from a sql database with pandas","text":"When manipulating you data using pandas, it is sometimes useful to pull data from a database. In this tutorial, we will see how to querry a dataframe from a sqlite table. Note than it would also work with any other sql database a long as you change the connxion to the one that suits your needs. First let's import pandas and sqlite3 import pandas as pd import sqlite3 from sqlite3 import Error We want to store the table tutyfrutty in our dataframe. To do so, we will query all the elements present in the tutyfrutty TABLE with the command : SELECT * FROM tutyfrutty db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : df = pd . read_sql ( \"SELECT * FROM tutyfrutty\" , conn ) del df [ \"index\" ] #juste delete the index column that was stored in the table except Error as e : print ( e ) df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 6 Cranberry red 308 7 Cranberry red 308","tags":"Python","url":"redoules.github.io/python/Reading_data_from_a_sql_database_with_pandas.html","loc":"redoules.github.io/python/Reading_data_from_a_sql_database_with_pandas.html"},{"title":"Writing data to a sql database with pandas","text":"When manipulating you data using pandas, it is sometimes useful to store a dataframe. Pandas provides multiple ways to export dataframes. The most common consist in exporting to a csv, a pickle, to hdf or to excel. However, exporting to a sql database can prove very useful. Indeed, having a well structured database is a great for storing all the data related to your analysis in one place. In this tutorial, we will see how to store a dataframe in a new table of a sqlite dataframe. Note than it would also work with any other sql database a long as you change the connxion to the one that suits your needs. First let's import pandas and sqlite3 import pandas as pd import sqlite3 from sqlite3 import Error # Example dataframe raw_data = { 'fruit' : [ 'Banana' , 'Orange' , 'Apple' , 'lemon' , \"lime\" , \"plum\" ], 'color' : [ 'yellow' , 'orange' , 'red' , 'yellow' , \"green\" , \"purple\" ], 'kcal' : [ 89 , 47 , 52 , 15 , 30 , 28 ] } df = pd . DataFrame ( raw_data , columns = [ 'fruit' , 'color' , 'kcal' ]) df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 Now that the DataFrame has been created, let's push it to the sqlite database called mydatabase.db in a new table called tutyfrutty db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : df . to_sql ( \"tutyfrutty\" , conn ) except Error as e : print ( e ) except ValueError : print ( \"The TABLE tutyfrutty already exists, read below to understand how to handle this case\" ) Note that if the table tutyfrutty was already existing, the to_sql function will return a ValueError. This is where, the if_exists option comes into play. Let's look at the docstring of this function : \"\"\" if_exists : {'fail', 'replace', 'append'}, default 'fail' - fail: If table exists, do nothing. - replace: If table exists, drop it, recreate it, and insert data. - append: If table exists, insert data. Create if does not exist. \"\"\" Let's say, I want to update my dataframe with some new rows df . loc [ len ( df ) + 1 ] = [ 'Cranberry' , 'red' , 308 ] df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 7 Cranberry red 308 8 Cranberry red 308 I can now replace the table with the new values using the \"replace\" option db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : df . to_sql ( \"tutyfrutty\" , conn , if_exists = \"replace\" ) except Error as e : print ( e )","tags":"Python","url":"redoules.github.io/python/Writing_data_to_a_sql_database_with_pandas.html","loc":"redoules.github.io/python/Writing_data_to_a_sql_database_with_pandas.html"},{"title":"Creating a sqlite database","text":"When you want to start with using databases SQlite is a great tool. It provides an easy onramp to learn and prototype you database with a SQL compatible database. First, let's import the libraries we need import sqlite3 from sqlite3 import Error SQlite doesn't need a database server, however, you have to start by creating an empty database file import os def check_for_db_file (): if os . path . exists ( \"mydatabase.db\" ): print ( \"the database is ready\" ) else : print ( \"no database found\" ) check_for_db_file () no database found Let's then create a function that will connect to a database, print the verison of sqlite and then close the connexion to the database. def create_database ( db_file ): \"\"\" create a database connection to a SQLite database \"\"\" try : with sqlite3 . connect ( db_file ) as conn : print ( \"database created with sqlite3 version {0}\" . format ( sqlite3 . version )) except Error as e : print ( e ) create_database ( \".\\mydatabase.db\" ) database created with sqlite3 version 2.6.0 check_for_db_file () the database is ready You're all set. From now on, you can open the database and write sql querries into it.","tags":"Python","url":"redoules.github.io/python/Creating_a_sqlite_database.html","loc":"redoules.github.io/python/Creating_a_sqlite_database.html"},{"title":"Setting up the notebook for plotting with matplotlib","text":"Importing Matplotlib First we need to import pyplot, a collection of command style functions that make matplotlib work like MATLAB. Let's, as well, use the magic command %matplotlib inline in order to display the figures in the notebook import matplotlib.pyplot as plt % matplotlib inline # this doubles image size, but we'll do it manually below # %config InlineBackend.figure_format = 'retina' The following parameters are recommended for matplotlib, they will make matplotlib output a better quality image # %load snippets/matplot_setup.py plt . rcParams [ 'savefig.dpi' ] = 300 plt . rcParams [ 'figure.dpi' ] = 163 plt . rcParams [ 'figure.autolayout' ] = False plt . rcParams [ 'figure.figsize' ] = 20 , 12 plt . rcParams [ 'axes.labelsize' ] = 18 plt . rcParams [ 'axes.titlesize' ] = 20 plt . rcParams [ 'font.size' ] = 16 plt . rcParams [ 'lines.linewidth' ] = 2.0 plt . rcParams [ 'lines.markersize' ] = 8 plt . rcParams [ 'legend.fontsize' ] = 14 plt . rcParams [ 'text.usetex' ] = False # True activates latex output in fonts! plt . rcParams [ 'font.family' ] = \"serif\" plt . rcParams [ 'font.serif' ] = \"cm\" plt . rcParams [ 'text.latex.preamble' ] = \" \\\\ usepackage{subdepth}, \\\\ usepackage{type1cm}\" You can change the second line in order to fit your display. 163 dpi corresponds to a Dell Ultra HD 4k P2715Q. You can check your screen's dpi count at http://dpi.lv/","tags":"Python","url":"redoules.github.io/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html","loc":"redoules.github.io/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html"},{"title":"Why using a blockchain is a bad idea for your business","text":"What having a blockchain implies? storage costs : everyone maintaining the ledger needs to store every transaction bandwith costs : everyone has to broadcast every transaction computational costs : every node has to validate the blockchain control : the creator does not control the blockchain, everyone collectively controls it developpement costs : developping on a blockchain is way harder than on a traditionnal database What to ask a business when they tell you that they are using a blockchain? When a business is telling you about their innovative technology leveraging the power of the blockchain this should immedialty spake some questions : What is the consensus algorithm? who is responsible for validating the consensus rules? what is the nature of the participation ? is it open to access? is it open to innovation? is it a public ledger? is it transparent? does it improves acountability? is it cross borders? how is it validated?","tags":"Cryptocurrencies","url":"redoules.github.io/cryptocurrencies/blockchain_bad.html","loc":"redoules.github.io/cryptocurrencies/blockchain_bad.html"},{"title":"Synology NFS share","text":"Setting up a NFS share login to your DSM admin account, open the \"Control Panel\" and go to \"File Services\" Make sure NFS is enabled Back in the control panel, go to \"Shared Folder\" Select the folder you want to share and clic \"Edit\" Go to the \"NFS Permissions tab and clic \"Create\", add the ip of the device you want to mount the mapped drive on. Make sure you copy the \"Mount path\"","tags":"Linux","url":"redoules.github.io/linux/share_nfs_share.html","loc":"redoules.github.io/linux/share_nfs_share.html"},{"title":"Mount a NFS share using fstab","text":"Mount nfs using fstab The fstab file, generally located at /etc/fstab lists the differents partitions and where to load them on the filesystem. You can edit this file as root by using the following command sudo nano /etc/fstab in the following example, we want to mount a NFS v3 share from : server : 192.168.1.2 mountpoint (on the server) : /volumeUSB2/usbshare * mountlocation (on the client) : /mnt we specify 192 .168.1.2:/volumeUSB2/usbshare /mnt nfs nfsvers = 3 ,users 0 0 the client will then automatically mount the share ont /mnt at startup. Related you can reload the fstab file using this method : https://redoules.github.io/linux/Reloading_fstab.html You can create a NFS share on a Synology using the method : https://redoules.github.io/linux/share_nfs_share.html","tags":"Linux","url":"redoules.github.io/linux/mount_nfs_share_fstab.html","loc":"redoules.github.io/linux/mount_nfs_share_fstab.html"},{"title":"Installing bitcoind on raspberry pi","text":"Installing bitcoind on linux Running a full bitcoin node helps the bitcoin network to accept, validate and relay transactions. If you want to volunteer some spare computing and bandwidth resources to run a full node and allow Bitcoin to continue to grow you can grab an inexpensive and power efficient raspberry pi and turn it into a full node. There are plenty of tutorials on the Internet explaining how to install a bitcoin full node; this tutorial won't go over setting up a raspberry pi and using ssh. In order to store the full blockchain we will mount a network drive and tell bitcoind to use this mapped drive as the data directory. Download the bitcoin client Go to https://bitcoin.org/en/download Copy the URL for the ARM 32 bit version and download it onto your raspberry pi. wget https://bitcoin.org/bin/bitcoin-core-0.15.1/bitcoin-0.15.1-arm-linux-gnueabihf.tar.gz Locate the downloaded file and extract it using the arguement xzf tar xzf bitcoin-0.15.1-arm-linux-gnueabihf.tar.gz a new directory bitcoin-0.15.1 will be created, it contrains the files we need to install the software Install the bitcoin client We will install the content by copying the binaries located in the bin folder into /usr/local/bin by using the install command. You must use sudo because it will write data to a system directory sudo install -m 0755 -o root -g root -t /usr/local/bin bitcoin-0.15.1/bin/* Launch the bitcoin core client by running bitcoind -daemon Configuration of the node Start your node at boot Starting you node automatically at boot time is a good idea because it doesn't require a manual action from the user. The simplest way to achive this is to create a cronjob. Run the following command crontab -e Select the text editor of your choice, then add the following line at the end of the file @reboot bitcoind -daemon Save the file and exit; the updated crontab file will be installed for you. Full Node If you can afford to download and store all the blockchain, you can run a full node. At the time of writing, the blockchain is 150Go ( https://blockchain.info/fr/charts/blocks-size ). Tree ways to store this are : use a microSD with 256Go or more add a thumbdrive or an external drive to your raspberry pi * mount a network drive from a NAS If you have purchased a big SD card then you can leave the default location for the blockchain data (~/.bitcoin/). Otherwise, you will have to change the datadir location to where your drive is mounted (in my case I have mounted it to /mnt) In order to configure your bitcoin client, edit/create the file bitcoin.conf located in ~/.bitcoin/ nano ~/.bitcoin/bitcoin.conf copy the following text # From redoules.github.io # This config should be placed in following path: # ~/.bitcoin/bitcoin.conf # [core] # Specify a non-default location to store blockchain and other data. datadir=/mnt # Set database cache size in megabytes; machines sync faster with a larger cache. Recommend setting as high as possible based upon mach$ dbcache=100 # Keep at most <n> unconnectable transactions in memory. maxorphantx=10 # Keep the transaction memory pool below <n> megabytess. maxmempool=50 # [network] # Maintain at most N connections to peers. maxconnections=40 # Tries to keep outbound traffic under the given target (in MiB per 24h), 0 = no limit. maxuploadtarget=5000 Check https://jlopp.github.io/bitcoin-core-config-generator it is a handy site to edit the bitcoin.conf file Pruning node If you don't want to store the entire blockchain you can run a pruning node which reduces storage requirements by enabling pruning (deleting) of old blocks. Let's say you want to allocated at most 5Go to the blockchain, then specify prune=5000 into your bitcoin.conf file. Edit/create the file bitcoin.conf located in ~/.bitcoin/ nano ~/.bitcoin/bitcoin.conf copy the following text # From redoules.github.io # This config should be placed in following path: # ~/.bitcoin/bitcoin.conf # [core] # Set database cache size in megabytes; machines sync faster with a larger cache. Recommend setting as high as possible based upon mach$ dbcache=100 # Keep at most <n> unconnectable transactions in memory. maxorphantx=10 # Keep the transaction memory pool below <n> megabytess. maxmempool=50 # Reduce storage requirements by only storing most recent N MiB of block. This mode is incompatible with -txindex and -rescan. WARNING: Reverting this setting requires re-downloading the entire blockchain. (default: 0 = disable pruning blocks, 1 = allow manual pruning via RPC, greater than 550 = automatically prune blocks to stay under target size in MiB). prune=5000 # [network] # Maintain at most N connections to peers. maxconnections=40 # Tries to keep outbound traffic under the given target (in MiB per 24h), 0 = no limit. maxuploadtarget=5000 Checking if your node is public one of the best way to help the bitcoin network is to allow your node to be visible and to propagate block to other nodes. The bitcoin protocole uses port 8333, other clients should be able to share information with your client. Run ifconfig and check if you have an ipv6 adresse (look for adr inet6:) IPV6 Get the global ipv6 adresse of your raspberry pi Link encap:Ethernet HWaddr xx:xx:xx:xx:xx:xx inet adr:192.168.1.x Bcast:192.168.1.255 Masque:255.255.255.0 adr inet6: xxxx::xxxx:xxxx:xxxx:xxxx/64 Scope:Lien adr inet6: xxxx:xxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:42681744 errors:0 dropped:0 overruns:0 frame:0 TX packets:38447218 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 lg file transmission:1000 RX bytes:3044414780 (2.8 GiB) TX bytes:2599878680 (2.4 GiB) it is located between adr inet4 and Scope:Global adr inet6: xxxx:xxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx/64 Scope:Global Copy this adresse and past it into the search field on https://bitnodes.earn.com/ If your node is visible, it will appear on the website IPV4 If you don't have an ipv6 adresse, you will have to open port 8333 on your router and redirect it to the internal IP of your raspberry pi. It is not detailed here because the configuration depends on your router.","tags":"Cryptocurrencies","url":"redoules.github.io/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html","loc":"redoules.github.io/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html"},{"title":"Reloading .bashrc","text":"Reload .bashrc The .bashrc file, located at ~/.bashrc allows a user to personalize its bash shell. If you edit this file, the changes won't be loaded without login out and back in. However, you can use the following command to do it source ~/.bashrc","tags":"Linux","url":"redoules.github.io/linux/Reloading_.bashrc.html","loc":"redoules.github.io/linux/Reloading_.bashrc.html"},{"title":"Reloading fstab","text":"Reload fstab The fstab file, generally located at /etc/fstab lists the differents partitions and where to load them on the filesystem. If you edit this file, the changes won't be automounted. You either have to reboot your system of use the following command as root mount -a","tags":"Linux","url":"redoules.github.io/linux/Reloading_fstab.html","loc":"redoules.github.io/linux/Reloading_fstab.html"},{"title":"Updating all python package with anaconda","text":"Updating anaconda packages All packages managed by conda can be updated with the following command : conda update --all Updating other packages with pip For the other packages, the pip package manager can be used. Unfortunately pip hasn't the same update all fonctionnality. import pip from subprocess import call for dist in pip . get_installed_distributions (): print ( \"updating {0}\" . format ( dist )) call ( \"pip install --upgrade \" + dist . project_name , shell = True )","tags":"Python","url":"redoules.github.io/python/updating_all_python_package_with_anaconda.html","loc":"redoules.github.io/python/updating_all_python_package_with_anaconda.html"},{"title":"Saving a matplotlib figure with a high resolution","text":"creating a matplotlib figure #Importing matplotlib % matplotlib inline import matplotlib.pyplot as plt import numpy as np Drawing a figure # Fixing random state for reproducibility np . random . seed ( 19680801 ) mu , sigma = 100 , 15 x = mu + sigma * np . random . randn ( 10000 ) # the histogram of the data n , bins , patches = plt . hist ( x , 50 , normed = 1 , facecolor = 'g' , alpha = 0.75 ) plt . xlabel ( 'Smarts' ) plt . ylabel ( 'Probability' ) plt . title ( 'Histogram of IQ' ) plt . text ( 60 , . 025 , r '$\\mu=100,\\ \\sigma=15$' ) plt . axis ([ 40 , 160 , 0 , 0.03 ]) plt . grid ( True ) plt . show () Saving the figure normally, one would use the following code plt . savefig ( 'filename.png' ) &lt;matplotlib.figure.Figure at 0x2e45e92f400&gt; The figure in then exported to the file \"filename.png\" with a standard resolution. In adittion, you can specify the dpi arg to some scalar value, for example: plt . savefig ( 'filename_hi_dpi.png' , dpi = 300 ) &lt;matplotlib.figure.Figure at 0x2e462164898&gt;","tags":"Python","url":"redoules.github.io/python/Saving_a_matplotlib_figure_with_a_high_resolution.html","loc":"redoules.github.io/python/Saving_a_matplotlib_figure_with_a_high_resolution.html"},{"title":"Iterating over a DataFrame","text":"Create a sample dataframe # Import modules import pandas as pd # Example dataframe raw_data = { 'fruit' : [ 'Banana' , 'Orange' , 'Apple' , 'lemon' , \"lime\" , \"plum\" ], 'color' : [ 'yellow' , 'orange' , 'red' , 'yellow' , \"green\" , \"purple\" ], 'kcal' : [ 89 , 47 , 52 , 15 , 30 , 28 ] } df = pd . DataFrame ( raw_data , columns = [ 'fruit' , 'color' , 'kcal' ]) df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 Using the iterrows method Pandas DataFrames can return a generator with the iterrrows method. It can then be used to loop over the rows of the DataFrame for index , row in df . iterrows (): print ( \"At line {0} there is a {1} which is {2} and contains {3} kcal\" . format ( index , row [ \"fruit\" ], row [ \"color\" ], row [ \"kcal\" ])) At line 0 there is a Banana which is yellow and contains 89 kcal At line 1 there is a Orange which is orange and contains 47 kcal At line 2 there is a Apple which is red and contains 52 kcal At line 3 there is a lemon which is yellow and contains 15 kcal At line 4 there is a lime which is green and contains 30 kcal At line 5 there is a plum which is purple and contains 28 kcal","tags":"Python","url":"redoules.github.io/python/Iterating_over_a_dataframe.html","loc":"redoules.github.io/python/Iterating_over_a_dataframe.html"},{"title":"Article Recommander","text":"import pandas as pd import numpy as np % matplotlib inline Loading data and preprocessing we first learn the pickled article database. We will be cleaning it and separating the interesting articles from the uninteresting ones. df = pd . read_pickle ( './article.pkl' ) del df [ \"html\" ] del df [ \"image\" ] del df [ \"URL\" ] del df [ \"hash\" ] del df [ \"source\" ] df [ \"label\" ] = df [ \"note\" ] . apply ( lambda x : 0 if x <= 0 else 1 ) df . head ( 5 ) .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } authors note resume texte titre label 0 [Danny Bradbury, Marco Santori, Adam Draper, M... -10.0 Black Market Reloaded, a black market site tha... Black Market Reloaded, a black market site tha... Black Market Reloaded back online after source... 0 1 [Emily Spaven, Stan Higgins, Emilyspaven] 1.0 The UK Home Office believes the government sho... The UK Home Office believes the government sho... Home Office: UK Should Create a Crime-Fighting... 1 2 [Pete Rizzo, Alex Batlin, Yessi Bello Perez, P... -10.0 Though lofty in its ideals, lead developer Dan... A new social messaging app is aiming to disrup... Gems Bitcoin App Lets Users Earn Money From So... 0 3 [Nermin Hajdarbegovic, Stan Higgins, Pete Rizz... 3.0 US satellite service provider DISH Network has... US satellite service provider DISH Network has... DISH Becomes World's Largest Company to Accept... 1 4 [Stan Higgins, Bailey Reutzel, Garrett Keirns,... -10.0 An unidentified 28-year-old man was robbed of ... An unidentified 28-year-old man was robbed of ... Bitcoin Stolen at Gunpoint in New York City Ro... 0 Basic statistics on the dataset let's explore the dataset and extract some numbers : * the number of article liked/disliked df [ \"label\" ] . value_counts () 0 879 1 324 Name: label, dtype: int64 Create the full content column df [ 'full_content' ] = df . titre + ' ' + df . resume #exclude the full texte of the article for the moment df . head ( 1 ) .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } authors note resume texte titre label full_content 0 [Danny Bradbury, Marco Santori, Adam Draper, M... -10.0 Black Market Reloaded, a black market site tha... Black Market Reloaded, a black market site tha... Black Market Reloaded back online after source... 0 Black Market Reloaded back online after source... from sklearn.model_selection import train_test_split training , testing = train_test_split ( df , # The dataset we want to split train_size = 0.75 , # The proportional size of our training set stratify = df . label , # The labels are used for stratification random_state = 400 # Use the same random state for reproducibility ) training . head ( 5 ) .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } authors note resume texte titre label full_content 748 [Jon Brodkin] -10.0 Amazon, Reddit, Mozilla, and other Internet co... Amazon, Reddit, Mozilla, and other Internet co... Amazon and Reddit try to save net neutrality r... 0 Amazon and Reddit try to save net neutrality r... 1183 [Jon Brodkin] -10.0 (The Time Warner involved in this transaction ... A group of mostly Democratic senators led by A... Democrats urge Trump administration to block A... 0 Democrats urge Trump administration to block A... 769 [Joseph Brogan] -10.0 On Twitter, bad news comes at all hours, with ... On Twitter, bad news comes at all hours, with ... Some of the best art on Twitter comes from the... 0 Some of the best art on Twitter comes from the... 57 [Michael Del Castillo, Pete Rizzo, Trond Vidar... -10.0 Publicly traded online travel service Webjet i... Publicly traded online travel service Webjet i... Webjet Ethereum Pilot Targets Hotel Industry's... 0 Webjet Ethereum Pilot Targets Hotel Industry's... 892 [Andrew Cunningham] 10.0 What has changed on the 2017 MacBook, then?\\nI... Andrew Cunningham\\n\\nAndrew Cunningham\\n\\nAndr... Mini-review: The 2017 MacBook could actually b... 1 Mini-review: The 2017 MacBook could actually b... from sklearn.feature_extraction.text import TfidfVectorizer , CountVectorizer from sklearn.svm import LinearSVC , SVC from sklearn.pipeline import Pipeline from sklearn.model_selection import cross_val_predict from utils.plotting import pipeline_performance steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , LinearSVC ()) ) pipeline = Pipeline ( steps ) predicted_labels = cross_val_predict ( pipeline , training . full_content , training . label ) pipeline_performance ( training . label , predicted_labels ) pipeline = pipeline . fit ( training . titre , training . label ) Accuracy = 80.6% Confusion matrix, without normalization [[624 35] [140 103]] import re from utils.plotting import print_top_features from sklearn.model_selection import GridSearchCV def mask_integers ( s ): return re . sub ( r '\\d+' , 'INTMASK' , s ) steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , LinearSVC ()) ) pipeline = Pipeline ( steps ) gs_params = { #'vectorizer__use_idf': (True, False), 'vectorizer__lowercase' : [ True , False ], 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__C' : np . linspace ( 5 , 20 , 25 ) } gs = GridSearchCV ( pipeline , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline1 = gs . best_estimator_ predicted_labels = pipeline1 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline1 , n_features = 10 ) aaa = gs . predict ( testing . full_content ) == testing . label aaa = aaa [ testing . label == 1 ] testing [ \"titre\" ] . iloc [ ~ aaa . values ] #pipeline1.predict([\"windows xbox bitcoin\"]) from sklearn.externals import joblib joblib . dump ( pipeline1 , 'classifier.pkl' ) gs . predict ([ 'Google' ]) array([1], dtype=int64) steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , SVC ()) ) pipeline = Pipeline ( steps ) gs_params = { #'vectorizer__use_idf': (True, False), 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__C' : np . linspace ( 5 , 20 , 25 ) } gs = GridSearchCV ( pipeline , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline1 = gs . best_estimator_ predicted_labels = pipeline1 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline1 , n_features = 10 ) {'classifier__C': 5.0, 'vectorizer__ngram_range': (1, 1), 'vectorizer__preprocessor': &lt;function mask_integers at 0x00000237491B67B8&gt;, 'vectorizer__stop_words': 'english'} 0.711180124224 Accuracy = 71.2% Confusion matrix, without normalization [[153 0] [ 62 0]] --------------------------------------------------------------------------- ValueError Traceback (most recent call last) &lt;ipython-input-9-3e0781e307fb&gt; in &lt;module&gt;() 25 pipeline_performance(testing.label, predicted_labels) 26 ---&gt; 27 print_top_features(pipeline1, n_features=10) C:\\Users\\Guillaume\\Documents\\Code\\recommandation\\utils\\plotting.py in print_top_features(pipeline, vectorizer_name, classifier_name, n_features) 81 def print_top_features(pipeline, vectorizer_name='vectorizer', classifier_name='classifier', n_features=7): 82 vocabulary = np.array(pipeline.named_steps[vectorizer_name].get_feature_names()) ---&gt; 83 coefs = pipeline.named_steps[classifier_name].coef_[0] 84 top_feature_idx = np.argsort(coefs) 85 top_features = vocabulary[top_feature_idx] C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\svm\\base.py in coef_(self) 483 def coef_(self): 484 if self.kernel != 'linear': --&gt; 485 raise ValueError('coef_ is only available when using a ' 486 'linear kernel') 487 ValueError: coef_ is only available when using a linear kernel from sklearn.naive_bayes import BernoulliNB steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , BernoulliNB ()) ) pipeline2 = Pipeline ( steps ) gs_params = { 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__alpha' : np . linspace ( 0 , 1 , 5 ), 'classifier__fit_prior' : [ True , False ] } gs = GridSearchCV ( pipeline2 , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline2 = gs . best_estimator_ predicted_labels = pipeline2 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline2 , n_features = 10 ) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - {'classifier__alpha': 0.25, 'classifier__fit_prior': True, 'vectorizer__ngram_range': (1, 1), 'vectorizer__preprocessor': &lt;function mask_integers at 0x00000237491B67B8&gt;, 'vectorizer__stop_words': 'english'} 0.805900621118 Accuracy = 78.1% Confusion matrix, without normalization [[140 13] [ 34 28]] Top like features: ['use' 'just' 'year' 'price' 'time' 'Bitcoin' 'bitcoin' 'new' 'The' 'INTMASK'] --- Top dislike features: ['ABBA' 'cable' 'cab' 'byte' 'publication' 'bye' 'publications' 'publicity' 'buyer' 'publicizing'] from sklearn.naive_bayes import MultinomialNB steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , MultinomialNB ()) ) pipeline3 = Pipeline ( steps ) gs_params = { 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__alpha' : np . linspace ( 0 , 1 , 5 ), 'classifier__fit_prior' : [ True , False ] } gs = GridSearchCV ( pipeline3 , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline3 = gs . best_estimator_ predicted_labels = pipeline3 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline3 , n_features = 10 ) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - {'classifier__alpha': 0.5, 'classifier__fit_prior': False, 'vectorizer__ngram_range': (1, 1), 'vectorizer__preprocessor': &lt;function mask_integers at 0x00000237491B67B8&gt;, 'vectorizer__stop_words': 'english'} 0.80900621118 Accuracy = 79.1% Confusion matrix, without normalization [[141 12] [ 33 29]] Top like features: ['time' 'Google' 'Pro' 'Apple' 'new' 'The' 'Bitcoin' 'price' 'bitcoin' 'INTMASK'] --- Top dislike features: ['ABBA' 'categories' 'catching' 'catalyst' 'catalog' 'casually' 'casts' 'cast' 'cashier' 'ran']","tags":"Machine Learning","url":"redoules.github.io/machine-learning/Source code for the recommandation engine for articles.html","loc":"redoules.github.io/machine-learning/Source code for the recommandation engine for articles.html"}]}
\ No newline at end of file
+{"pages":[{"title":"About Guillaume Redoulès","text":"I am a data scientist and a mechanical engineer working on numerical methods for stress computations in the field of rocket propulsion. Prior to that, I've got a MSc in Computational Fluid Dynamics and aerodynamics from Imperial College London. Email: guillaume.redoules@gadz.org Linkedin: Guillaume Redoulès Curriculum Vitae Experience Thermomecanical method and tools engineer , Ariane Group , 2015 - Present In charge of tools and methods related to thermomecanical computations. Focal point for machine learning. Education MSc Advanced Computational Methods for Aeronautics, Flow Management and Fluid-Structure Interaction , Imperial College London, London. 2013 Dissertation: \"Estimator design for fluid flows\" Fields: Aeronautics, aerodynamics, computational fluid dynamics, numerical methods Arts et Métiers Paristech , France, 2011 Generalist engineering degree Fields: Mechanics, electrical engineering, casting, machining, project management, finance, IT, etc.","tags":"pages","url":"redoules.github.io/pages/about.html","loc":"redoules.github.io/pages/about.html"},{"title":"Create a simple bash function","text":"A basic function The synthaxe to define a function is : #!/bin/bash # Basic function my_function () { echo Text displayed by my_function } #once defined, you can use it like so : my_function and it should return user@bash : ./my_function.sh Text displayed by my_function Function with arguments When used, the arguments are specified directly after the function name. Whithin the function they are accessible this the $ symbol followed by the number of the arguement. Hence $1 will take the value of the first arguement, $2 will take the value of the second arguement and so on. #!/bin/bash # Passing arguments to a function say_hello () { echo Hello $1 } say_hello Guillaume and it should return user@bash : ./function_arguements.sh Hello Guillaume Overriding Commands Using the previous example, let's override the echo function in order to make it say hello. To do so, you just need to name the function with the same name as the command you want to replace. When you are calling the original function, make sure you are using the builtin keyword #!/bin/bash # Overriding a function echo () { builtin echo Hello $1 } echo Guillaume user@bash : ./function_arguements.sh Hello Guillaume Returning values Use the keyword return to send back a value to the main program. The returned value will be stored in the $? variable #!/bin/bash # Retruning a value secret_number () { return 126 } secret_number echo The secret number is $? This code should return user@bash : ./retrun_value.sh The secret number is 126","tags":"Linux","url":"redoules.github.io/linux/simple_bash_function.html","loc":"redoules.github.io/linux/simple_bash_function.html"},{"title":"Number of edges in a Complete graph","text":"A complete graph contains $\\frac{n(n-1)}{2}$ edges where $n$ is the number of vertices (or nodes).","tags":"Mathematics","url":"redoules.github.io/mathematics/Number_edges_Complete_graph.html","loc":"redoules.github.io/mathematics/Number_edges_Complete_graph.html"},{"title":"Reverse an array","text":"Using lists Python provides an iterator to reverse an array reversed() you can use it this way : arr = list ( range ( 5 )) print ( f 'Initial array : {arr} \\n ' ) reversed_arr = list ( reversed ( arr )) print ( f 'Reversed array : {reversed_arr}' ) Initial array : [0, 1, 2, 3, 4] Reversed array : [4, 3, 2, 1, 0] Using numpy If you are using numpy you can reverse an array by creating a view on the array: import numpy as np arr = np . arange ( 5 ) print ( f 'Initial array : {arr} \\n ' ) reversed_arr = arr [:: - 1 ] print ( f 'Reversed array : {reversed_arr}' ) Initial array : [0 1 2 3 4] Reversed array : [4 3 2 1 0]","tags":"Python","url":"redoules.github.io/python/reverse.html","loc":"redoules.github.io/python/reverse.html"},{"title":"Advice for designing your own libraries","text":"Advice for designing your own libraries When designing your own library make sure to think of the following things. I will add new paragraphs to this article as I dicover new good practices. Use standard python objects Try to use standard python objects as much as possible. That way, your library becomes compatible with all the other python libaries. For instance, when I created SAMpy : a library for reading and writing SAMCEF results, it returned dictonnaries, lists and pandas dataframes. Hence the results extracted from SAMCEF where compatible with all the scientific stack of python. Limit the number of functionnalities Following the same logic as before, the objects should do only one thing but do it well. Indeed, having a simple interface will reduce the complexity of your code and make it easier to use your library. Again, with SAMpy, I decided to strictly limit the functionnalities to reading and writing SAMCEF files. Define an exception class for your library You should define your own exceptions in order to make it easier for your users to debug their code thanks to clearer messages that convey more meaning. That way, the user will know if the error comes from your library or something else. Bonus if you group similar exceptions in a hierachy of inerited Exception classes. Example : let's create a Exception related to the age of a person : def check_age ( age ): if age < 0 and age > 130 : raise ValueError If the user inputed an invalid age, the ValueError exception would be thrown. That's fine but imagine you wan't to provide more feedback to your users that don't know the internal of your library. Let's now create a selfexplanatory Exception class AgeInvalidError ( ValueError ): pass def check_age ( age ): if age < 0 and age > 130 : raise AgeInvalidError ( age ) You can also add some helpful text to guide your users along the way: class AgeInvalidError ( ValueError ): print ( \"Age invalid, must be between 0 and 130\" ) pass def check_age ( age ): if age < 0 and age > 130 : raise AgeInvalidError ( age ) If you want to group all the logically linked exceptions, you can create a base class and inherit from it : class BaseAgeInvalidError ( ValueError ): pass class TooYoungError ( BaseAgeInvalidError ): pass class TooOldError ( BaseAgeInvalidError ): pass def check_age ( age ): if age < 0 : raise TooYoungError ( age ) elif age > 130 : raise TooOldError ( age ) Structure your repository You should have a file structure in your repository. It will help other contributers especially future contributers. A nice directory structure for your project should look like this: README.md LICENSE setup.py requirements.txt ./MyPackage ./docs ./tests Some prefer to use reStructured Text, I personnaly prefer Markdown choosealicense.com will help you pick the license to use for your project. For package and distribution management, create a setup.py file a the root of the directory The list of dependencies required to test, build and generate the doc are listed in a pip requirement file placed a the root of the directory and named requirements.txt Put the documentation of your library in the docs directory. Put your tests in the tests directory. Since your tests will need to import your library, I recommend modifying the path to resolve your package property. In order to do so, you can create a context.py file located in the tests directory : import os import sys sys . path . insert ( 0 , os . path . abspath ( os . path . join ( os . path . dirname ( __file__ ), '..' ))) import MyPackage Then within your individual test files you can import your package like so : from .context import MyPackage Finally, your code will go into the MyPackage directory Test your code Once your library is in production, you have to guaranty some level of forward compatibility. Once your interface is defined, write some tests. In the future, when your code is modified, having those tests will make sure that the behaviour of your functions and objects won't be altered. Document your code Of course, you should have a documentation to go along with your library. Make sure to add a lot of commun examples as most users tend to learn from examples. I recommend writing your documentation using Sphinx.","tags":"Python","url":"redoules.github.io/python/design_own_libs.html","loc":"redoules.github.io/python/design_own_libs.html"},{"title":"Safely creating a folder if it doesn't exist","text":"Safely creating a folder if it doesn't exist When you are writing to files in python, if the file doesn't exist it will be created. However, if you are trying to write a file in a directory that doesn't exist, an exception will be returned FileNotFoundError : [ Errno 2 ] No such file or directory : \"directory\" This article will teach you how to make sure the target directory exists. If it doesn't, the function will create that directory. First, let's import os and make sure that the \"test_directory\" doesn't exist import os os . path . exists ( \". \\\\ test_directory\" ) False copy the ensure_dir function in your code. This function will handle the creation of the directory. Credit goes to Parand posted on StackOverflow def ensure_dir ( file_path ): directory = os . path . dirname ( file_path ) if not os . path . exists ( directory ): os . makedirs ( directory ) Let's now use the function and create a folder named \"test_directory\" ensure_dir ( \". \\\\ test_directory\" ) If we test for the existence of the directory, the exists function will now return True os . path . exists ( \". \\\\ test_directory\" ) True","tags":"Python","url":"redoules.github.io/python/ensure_dir.html","loc":"redoules.github.io/python/ensure_dir.html"},{"title":"List all files in a directory","text":"Listing all the files in a directory Let's start with the basics, the most staigthforward way to list all the files in a direcoty is to use a combinaison of the listdir function and isfile form os.path. You can use a list comprehension to store all the results in a list. mypath = \"./test_directory/\" from os import listdir from os.path import isfile , join [ f for f in listdir ( mypath ) if isfile ( join ( mypath , f ))] ['logfile.log', 'myfile.txt', 'super_music.mp3', 'textfile.txt'] Listing all the files of a certain type in a directory similarly, if you want to filter only a certain kind of file based on its extension you can use the endswith method. In the following example, we will filter all the \"txt\" files contained in the directory [ f for f in listdir ( mypath ) if f . endswith ( '.' + \"txt\" )] ['myfile.txt', 'textfile.txt'] Listing all the files matching a pattern in a directory The glob module finds all the pathnames matching a specified pattern according to the rules used by the Unix shell. You can use the *, ?, and character ranges expressed with [] wildcards import glob glob . glob ( \"*.txt\" ) ['myfile.txt'] Listing files recusively If you want to list all files recursively you can select all the sub-directories using the \"**\" wildcard import glob glob . glob ( mypath + '/**/*.txt' , recursive = True ) ['./test_directory\\\\myfile.txt', './test_directory\\\\textfile.txt', './test_directory\\\\subdir1\\\\file_hidden_in_a_sub_direcotry.txt'] Using a regular expression If you'd rather use a regular expression to select the files, the pathlib library provides the rglob function. from pathlib import Path list ( Path ( \"./test_directory/\" ) . rglob ( \"*.[tT][xX][tT]\" )) [WindowsPath('test_directory/myfile.txt'), WindowsPath('test_directory/textfile.txt'), WindowsPath('test_directory/subdir1/file_hidden_in_a_sub_direcotry.txt')] Using regular expressions you can for example select multiple types of files. In the following example, we list all the files that finish either with \"txt\" or with \"log\". list ( Path ( \"./test_directory/\" ) . rglob ( \"*.[tl][xo][tg]\" )) [WindowsPath('test_directory/logfile.log'), WindowsPath('test_directory/myfile.txt'), WindowsPath('test_directory/textfile.txt'), WindowsPath('test_directory/subdir1/file_hidden_in_a_sub_direcotry.txt')]","tags":"Python","url":"redoules.github.io/python/list_files_directory.html","loc":"redoules.github.io/python/list_files_directory.html"},{"title":"Using Dask on infiniband","text":"InfiniBand (abbreviated IB) is a computer-networking communications standard used in high-performance computing that features very high throughput and very low latency. It is used for data interconnect both among and within computers. InfiniBand is also used as either a direct or switched interconnect between servers and storage systems, as well as an interconnect between storage systems. (source Wikipedia). If you want to leverage this high speed network instead of the regular ethernet network, you have to specify to the scheduler that you want to used infiniband as your interface. Assuming that you Infiniband interface is ib0 , you would call the scheduler like this : dask-scheduler --interface ib0 --scheduler-file ./cluster.yaml you would have to call the worker using the same interface : dask-worker --interface ib0 --scheduler-file ./cluster.yaml","tags":"Python","url":"redoules.github.io/python/dask_infiniband.html","loc":"redoules.github.io/python/dask_infiniband.html"},{"title":"Clearing the current cell in the notebook","text":"In python, you can clear the output of a cell by importing the IPython.display module and using the clear_output function from IPython.display import clear_output print ( \"text to be cleared\" ) clear_output () As you can see, the text \"text to be cleared\" is not displayed because the function clear_output has been called afterward","tags":"Jupyter","url":"redoules.github.io/jupyter/clear_cell.html","loc":"redoules.github.io/jupyter/clear_cell.html"},{"title":"What's inside my .bashrc ?","text":"############ # Anaconda # ############ export PATH = \"/station/guillaume/anaconda3/bin: $PATH \" alias python = '/station/guillaume/anaconda3/bin/python' ######### # Alias # ######### alias ding = 'echo -e \"\\a\"' alias calc = 'python -ic \"from __future__ import division; from math import *\"' alias h = \"history|grep \" alias f = \"find . |grep \" alias p = \"ps aux |grep \" alias cdl = \"cd /data/guillaume\" alias cp = \"rsync -avz --progress\" alias grep = \"grep --color=auto\" alias ls = \"ls -hN --color=auto --group-directories-first\" alias ll = \"ls -hal\" alias sv = \"ssh compute_cluster\" alias ms = \"ls\" alias jl = \"jupyter lab\" alias lst = \"jupyter notebook list\" ########################## # bashrc personnalisation# ########################## force_color_prompt = yes export EDITOR = nano export BROWSER = \"firefox '%s' &\" if [ -n \" $force_color_prompt \" ] ; then if [ -x /usr/bin/tput ] && tput setaf 1 > & /dev/null ; then # We have color support; assume it's compliant with Ecma-48 # (ISO/IEC-6429). (Lack of such support is extremely rare, and such # a case would tend to support setf rather than setaf.) color_prompt = yes else color_prompt = fi fi if [ \" $color_prompt \" = yes ] ; then #\\h : hostname #\\u : user #\\w : current working directory #\\d : date #\\t : time yellow = 226 green = 83 pink = 198 blue = 34 PS1 = \"\\[\\033[38;5;22m\\]\\u\\[ $( tput sgr0 ) \\]\\[\\033[38;5;163m\\]@\\[ $( tput sgr0 ) \\]\\[\\033[38;5;22m\\]\\h\\[ $( tput sgr0 ) \\]\\[\\033[38;5;162m\\]:\\[ $( tput sgr0 ) \\]\\[\\033[38;5;172m\\]{\\[ $( tput sgr0 ) \\]\\[\\033[38;5;39m\\]\\w\\[ $( tput sgr0 ) \\]\\[\\033[38;5;172m\\]}\\[ $( tput sgr0 ) \\]\\[\\033[38;5;162m\\]>\\[ $( tput sgr0 ) \\]\"","tags":"Linux","url":"redoules.github.io/linux/bashrc.html","loc":"redoules.github.io/linux/bashrc.html"},{"title":"Efficient extraction of eigenvalues from a list of tensors","text":"When you manipulate FEM results you generally have either a: scalar field, vector field, * tensor field. With tensorial results, it is often useful to extract the eigenvalues in order to find the principal values. I have found that it is easier to store the components of the tensors in a 6 column pandas dataframe (because of the symmetric property of stress and strain tensors) import pandas as pd node = [ 1001 , 1002 , 1003 , 1004 ] #when dealing with FEM results you should remember at which element/node the result is computed (in the example, let's assume that we look at node from 1001 to 1004) tensor1 = [ 1 , 1 , 1 , 0 , 0 , 0 ] #eigen : 1 tensor2 = [ 4 , - 1 , 0 , 2 , 2 , 1 ] #eigen : 5.58443, -1.77931, -0.805118 tensor3 = [ 1 , 6 , 5 , 3 , 3 , 1 ] #eigen : 8.85036, 4.46542, -1.31577 tensor4 = [ 1 , 2 , 3 , 0 , 0 , 0 ] #eigen : 1, 2, 3 df = pd . DataFrame ([ tensor1 , tensor2 , tensor3 , tensor4 ], columns = [ \"XX\" , \"YY\" , \"ZZ\" , \"XY\" , \"XZ\" , \"YZ\" ]) df . index = node df .dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } XX YY ZZ XY XZ YZ 1001 1 1 1 0 0 0 1002 4 -1 0 2 2 1 1003 1 6 5 3 3 1 1004 1 2 3 0 0 0 If you want to extract the eigenvalues of a tensor with numpy you have to pass a n by n ndarray to the eigenvalue function. In order to avoid having to loop over each node, this oneliner is highly optimized and will help you invert a large number of tensors efficiently. The steps are basically, create a list of n by n values (here n=3) in the right order => reshape it to a list of tensors => pass it to the eigenvals function import numpy as np from numpy import linalg as LA eigenvals = LA . eigvals ( df [[ \"XX\" , \"XY\" , \"XZ\" , \"XY\" , \"YY\" , \"YZ\" , \"XZ\" , \"YZ\" , \"ZZ\" ]] . values . reshape ( len ( df ), 3 , 3 )) eigenvals array([[ 1. , 1. , 1. ], [ 5.58442834, -0.80511809, -1.77931025], [-1.31577211, 8.85035616, 4.46541595], [ 1. , 2. , 3. ]])","tags":"Python","url":"redoules.github.io/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html","loc":"redoules.github.io/python/Efficient_extraction_of_eigenvalues_from_a_list_of_tensors.html"},{"title":"Optimized numpy random number generation on Intel CPU","text":"Python Intel distribution Make sure you have a python intel distribution. When you startup python you should see somethine like : Python 3.6.2 |Intel Corporation| (default, Aug 15 2017, 11:34:02) [MSC v.1900 64 bit (AMD64)] Type 'copyright', 'credits' or 'license' for more information IPython 6.1.0 -- An enhanced Interactive Python. Type '?' for help. If not, you can force the installation of the intel optimized python with : conda update --all conda config --add channels intel conda install numpy --channel intel --override-channels oh and by the way, make sure you a running an Intel CPU ;) Comparing numpy.random with numpy.random_intel Let's now test both the rand function with and without the Intel optimization import numpy as np from numpy import random , random_intel % timeit np . random . rand ( 10 ** 5 ) 1.06 ms ± 91.5 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each) % timeit np . random_intel . rand ( 10 ** 5 ) 225 µs ± 3.46 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)","tags":"Python","url":"redoules.github.io/python/Optimized_numpy_random_intel.html","loc":"redoules.github.io/python/Optimized_numpy_random_intel.html"},{"title":"How to check Linux process information?","text":"How to check Linux process information (CPU usage, memory, user information, etc.)? You need to use the ps command combined with the grep command. In the example, we want to check the information on the nginx process : ps aux | grep nginx It would return the output : root 9976 0.0 0.0 12272 108 ? S<s Aug12 0:00 nginx: master process /usr/bin/nginx -g pid /run/nginx.pid; daemon on; master_process on; http 16780 0.0 0.0 12384 684 ? S< Aug12 4:11 nginx: worker process http 16781 0.0 0.0 12556 708 ? S< Aug12 0:24 nginx: worker process http 16782 0.0 0.1 12292 744 ? S< Aug12 2:43 nginx: worker process http 16783 0.0 0.1 12276 872 ? S< Aug12 0:24 nginx: worker process admin 17612 0.0 0.1 5120 864 pts/4 S+ 11:22 0:00 grep --color=auto nginx The columns have the following order : USER;PID;%CPU;%MEM;VSZ;RSS;TTY;STAT;START;TIME;COMMAND USER = user owning the process PID = process ID of the process %CPU = It is the CPU time used divided by the time the process has been running. %MEM = ratio of the process's resident set size to the physical memory on the machine VSZ = virtual memory usage of entire process (in KiB) RSS = resident set size, the non-swapped physical memory that a task has used (in KiB) TTY = controlling tty (terminal) STAT = multi-character process state START = starting time or date of the process TIME = cumulative CPU time COMMAND = command with all its arguments Interactive display If you want an interactive display showing in real time the statistics of the running process you can use the top command. If htop is available on you system, use this instead.","tags":"Linux","url":"redoules.github.io/linux/linux_process_information.html","loc":"redoules.github.io/linux/linux_process_information.html"},{"title":"Check the size of a directory","text":"How do you Check the size of a directory in linux? The du command will come handy for this task. Let's say we want to know the size of the directory named recommandations , we would run the following command du -sh recommendations It would return the output : 9.9M recommendations","tags":"Linux","url":"redoules.github.io/linux/directory_size.html","loc":"redoules.github.io/linux/directory_size.html"},{"title":"Check for free disk space","text":"How do you check for free disk space on linux? The df command will come handy for this task. Run the command with the following arguments df -ah Will return in a human readable format for all drives a readout of all your filesystems Filesystem Size Used Avail Use% Mounted on /dev/md0 2.4G 1.3G 1.1G 54% / none 348M 4.0K 348M 1% /dev none 0 0 0 - /dev/pts none 0 0 0 - /proc none 0 0 0 - /sys /tmp 350M 1.3M 349M 1% /tmp /run 350M 3.2M 347M 1% /run /dev/shm 350M 12K 350M 1% /dev/shm /proc/bus/usb 0 0 0 - /proc/bus/usb securityfs 0 0 0 - /sys/kernel/security /dev/md3 1.8T 372G 1.5T 21% /volume2 /dev/vg1000/lv 1.8T 1.5T 340G 82% /volume1 /dev/sdq1 7.4G 3.8G 3.5G 52% /volumeUSB3/usbshare3-1 /dev/sdr 294G 146G 134G 53% /volumeUSB2/usbshare none 0 0 0 - /proc/fs/nfsd none 0 0 0 - /config The free disk space can be read in the Avail column","tags":"Linux","url":"redoules.github.io/linux/free_disk_space_linux.html","loc":"redoules.github.io/linux/free_disk_space_linux.html"},{"title":"Check your current ip address","text":"Check your current ip address Run the command ip addr show and that will give you every information available 4: eth0: <> mtu 1500 group default qlen 1 link/ether 5c:51:4f:41:7a:b1 inet 169.254.33.33/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::390a:f69e:1ba2:2121/64 scope global dynamic valid_lft forever preferred_lft forever 3: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 group default qlen 1 link/ether 0a:00:27:00:00:03 inet 192.168.56.1/24 brd 192.168.56.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::84cd:374e:843f:f82f/64 scope global dynamic valid_lft forever preferred_lft forever 15: eth2: <> mtu 1500 group default qlen 1 link/ether 00:ff:d2:8a:19:c3 inet 169.254.40.62/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::f199:2ca0:7aff:283e/64 scope global dynamic valid_lft forever preferred_lft forever 1: lo: <LOOPBACK,UP> mtu 1500 group default qlen 1 link/loopback 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 ::1/128 scope global dynamic valid_lft forever preferred_lft forever 5: wifi0: <BROADCAST,MULTICAST,UP> mtu 1500 group default qlen 1 link/ieee802.11 5c:51:4f:41:7a:ad inet 192.168.1.1/24 brd 192.168.1.255 scope global dynamic valid_lft 42720sec preferred_lft 42720sec inet6 fe80::395f:3594:1dc2:57e3/64 scope global dynamic valid_lft forever preferred_lft forever 21: wifi1: <> mtu 1500 group default qlen 1 link/ieee802.11 5c:51:4f:41:7a:ae inet 169.254.12.77/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::58d5:630:cbbd:c4d/64 scope global dynamic valid_lft forever preferred_lft forever 12: eth3: <> mtu 1472 group default qlen 1 link/ether 00:00:00:00:00:00:00:e0:00:00:00:00:00:00:00:00 inet6 fe80::100:7f:fffe/64 scope global dynamic valid_lft forever preferred_lft forever 10: eth4: <BROADCAST,MULTICAST,UP> mtu 1500 group default qlen 1 link/ether 22:b7:57:52:5f:ff inet 192.168.42.106/24 brd 192.168.42.255 scope global dynamic valid_lft 6659sec preferred_lft 6659sec inet6 fe80::5110:eb6f:deb0:45c4/64 scope global dynamic valid_lft forever preferred_lft forever You can select only one interface ip addr show eth0 and only the relevant information will be displayed 4: eth0: <> mtu 1500 group default qlen 1 link/ether 5c:51:4f:41:7a:b1 inet 169.254.33.33/16 brd 169.254.255.255 scope global dynamic valid_lft forever preferred_lft forever inet6 fe80::390a:f69e:1ba2:2121/64 scope global dynamic valid_lft forever preferred_lft forever Check your current ip address the old way The ifconfig command will return information regarding your network interfaces. Let's try it: ifconfig eth1 Link encap:Ethernet HWaddr 0a:00:27:00:00:03 inet adr:192.168.56.1 Bcast:192.168.56.255 Masque:255.255.255.0 adr inet6: fe80::84cd:374e:843f:f82f/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) eth4 Link encap:Ethernet HWaddr 22:b7:57:52:5f:ff inet adr:192.168.42.106 Bcast:192.168.42.255 Masque:255.255.255.0 adr inet6: fe80::5110:eb6f:deb0:45c4/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) lo Link encap:Boucle locale inet adr:127.0.0.1 Masque:255.0.0.0 adr inet6: ::1/128 Scope:Global UP LOOPBACK RUNNING MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) wifi0 Link encap:UNSPEC HWaddr 5C-51-4F-41-7A-AD-00-00-00-00-00-00-00-00-00-00 inet adr:192.168.1.1 Bcast:192.168.1.255 Masque:255.255.255.0 adr inet6: fe80::395f:3594:1dc2:57e3/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 Packets reçus:0 erreurs:0 :0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 Octets reçus:0 (0.0 B) Octets transmis:0 (0.0 B) On the left column you have the list of network adapters. The lo is the local loopback, it is an interface that points to the localhost. Interfaces starting with eth refer to wired connections over ethernet (or sometimes USB in the case of a phone acting as an access point over USB). Interfaces starting with wlan or wifi refer to wireless connections. On the right column you have some information corresponding to the interface such as the IPv4, the IPv6, the mask, some statistics about the interface and so on.","tags":"Linux","url":"redoules.github.io/linux/get_ip_linux.html","loc":"redoules.github.io/linux/get_ip_linux.html"},{"title":"Check the version of the kernel currently running","text":"Check the version of the kernel currently running The uname command will give you the version of the kernel. In order to get a more useful output, type uname -a This will return : the hostname os name kernel release/version architecture * etc. variations If you only want the kernel version you can type uname -v if you only want the kernel release you can type uname -r","tags":"Linux","url":"redoules.github.io/linux/version_kernel.html","loc":"redoules.github.io/linux/version_kernel.html"},{"title":"Running the notebook on a remote server","text":"Jupyter hub With JupyterHub you can create a multi-user Hub which spawns, manages, and proxies multiple instances of the single-user Jupyter notebook server. Project Jupyter created JupyterHub to support many users. The Hub can offer notebook servers to a class of students, a corporate data science workgroup, a scientific research project, or a high performance computing group. However, if you are the only one using the server and you just want a simple way to run the notebook on your server and access it through the web interface on a light client without having to install and configure the jupyter hub, you can do the following. Problem with jupyter notebook On your server, run the command jupyter-notebook you should get something like : [I 11:18:44.514 NotebookApp] Serving notebooks from local directory: /volume2/homes/admin [I 11:18:44.515 NotebookApp] 0 active kernels [I 11:18:44.516 NotebookApp] The Jupyter Notebook is running at: http://localhost:8888/?token=357587e3269b0f20f2b7e1918492890ae7573ac7ef1d2023 [I 11:18:44.516 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). [W 11:18:44.519 NotebookApp] No web browser found: could not locate runnable browser. [C 11:18:44.520 NotebookApp] Copy/paste this URL into your browser when you connect for the first time, to login with a token: http://localhost:8888/?token=357587e3269b0f20f2b7e1918492890ae7573ac7ef1d2023 and if you try to connect to your server ip (in my example : http://192.168.1.2:8888/?token=357587e3269b0f20f2b7e1918492890ae7573ac7ef1d2023) you will get an \"ERR_CONNECTION_REFUSED\" error. This is because, by default, Jupyter Notebook only accepts connections from localhost. Allowing connexions from other sources From any IP The simplest way to avoid the connection error is to allow the notebook to accept connections from any ip jupyter-notebook --ip = * you will get something like [W 11:26:45.285 NotebookApp] WARNING: The notebook server is listening on all IP addresses and not using encryption. This is not recommended. [I 11:26:45.342 NotebookApp] Serving notebooks from local directory: /volume2/homes/admin [I 11:26:45.342 NotebookApp] 0 active kernels [I 11:26:45.343 NotebookApp] The Jupyter Notebook is running at: http://[all ip addresses on your system]:8888/?token=52af33d628881824968b4031967e8541a27cc28b1720c199 [I 11:26:45.343 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). [W 11:26:45.346 NotebookApp] No web browser found: could not locate runnable browser. [C 11:26:45.347 NotebookApp] Copy/paste this URL into your browser when you connect for the first time, to login with a token: http://localhost:8888/?token=52af33d628881824968b4031967e8541a27cc28b1720c199 and if you connect form a remote client (192.168.1.1 in my example), the following line will be added to the output : [I 11:26:54.798 NotebookApp] 302 GET /?token=52af33d628881824968b4031967e8541a27cc28b1720c199 (192.168.1.1) 111.17ms note that you should only do that if you are the only one using the server because the connection is not encypted. From a specific IP You can also, explicitly specify the ip of the client jupyter-notebook --ip = 192 .168.1.1 [I 11:44:58.104 NotebookApp] JupyterLab extension loaded from C:\\Users\\Guillaume\\Miniconda3\\lib\\site-packages\\jupyterlab [I 11:44:58.104 NotebookApp] JupyterLab application directory is C:\\Users\\Guillaume\\Miniconda3\\share\\jupyter\\lab [I 11:44:58.244 NotebookApp] Serving notebooks from local directory: C:\\Users\\Guillaume [I 11:44:58.245 NotebookApp] 0 active kernels [I 11:44:58.245 NotebookApp] The Jupyter Notebook is running at: http://192.168.1.1:8888/?token=503576dd8fa87d1f2c416df307e9b900e520b4942e317b32 [I 11:44:58.245 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation). [C 11:44:58.258 NotebookApp] Copy/paste this URL into your browser when you connect for the first time, to login with a token: http://192.168.1.1:8888/?token=503576dd8fa87d1f2c416df307e9b900e520b4942e317b32 [I 11:44:59.083 NotebookApp] Accepting one-time-token-authenticated connection from 192.168.1.1","tags":"Jupyter","url":"redoules.github.io/jupyter/remote_run_notebook.html","loc":"redoules.github.io/jupyter/remote_run_notebook.html"},{"title":"Running multiple calls to a function in parallel with Dask","text":"Dask.distributed is a lightweight library for distributed computing in Python. It allows to create a compute graph. Dask distributed is architectured around 3 parts : the dask-scheduler the dask-worker(s) the dask client Dask architecture The Dask scheduler is a centrally managed, distributed, dynamic task scheduler. It recieves tasks from a/multiple client(s) and spread them across one or multiple dask-worker(s). Dask-scheduler is an event based asynchronous dynamic scheduler, meaning that mutliple clients can submit a list of task to be executed on multiple workers. Internally, the task are represented as a directed acyclic graph. Both new clients and new workers can be connected or disconnected during the execution of the task graph. Tasks can be submited with the function client . submit ( function , * args , ** kwargs ) or by using objects from the dask library such as dask.dataframe, dask.bag or dask.array Setup In this example, we will use a distributed scheduler on a single machine with multiple workers and a single client. We will use the client to submit some tasks to the scheduler. The scheduler will then dispatch those tasks to the workers. The process can be monitored in real time through a web application. For this example, all the computations will be run on a local computer. However dask can scale to a large HPC cluster. First we have to launch the dask-scheduler; from the command line, input dask-scheduler Next, you can load the web dashboard. In order to do so, the scheduler returns the number of the port you have to connect to in the line starting with \"bokeh at :\". The default port is 8787. Since we are running all the programs on the same computer, we just have to login to http://127.0.0.1:8787/status Finally, we have to launch the dask-worker(s). If you want to run the worker(s) on the same computer as the scheduler the type : dask-worker 127 .0.0.1:8786 otherwise, make sure you are inputing the ip address of the computer hosting the dask-scheduler. You can launch as many workers as you want. In this example, we will run 3 workers on the local machine. Use the dask workers within your python code We will now see how to submit multiple calls to a fucntion in parallel on the dask-workers. Import the required libraries and define the function to be executed. import numpy as np import pandas as pd from distributed import Client #function used to do parallel computing on def compute_pi_MonteCarlo ( Nb_Data ): \"\"\" computes the value of pi using the monte carlo method \"\"\" Radius = 1 Nb_Data = int ( round ( Nb_Data )) x = np . random . uniform ( - Radius , Radius , Nb_Data ) y = np . random . uniform ( - Radius , Radius , Nb_Data ) pi_mc = 4 * np . sum ( np . power ( x , 2 ) + np . power ( y , 2 ) < Radius ** 2 ) / Nb_Data err = 100 * np . abs ( pi_mc - np . pi ) / np . pi return [ Nb_Data , pi_mc , err ] In order to connect to the scheduler, we create a client. client = Client ( '127.0.0.1:8786' ) client Client Scheduler: tcp://127.0.0.1:8786 Dashboard: http://127.0.0.1:8787/status Cluster Workers: 3 Cores: 12 Memory: 25.48 GB We submit tasks using the submit method data = [ client . submit ( compute_pi_MonteCarlo , Nb_Data ) for Nb_Data in np . logspace ( 3 , 7 , num = 1200 , dtype = int )] If you look at http://127.0.0.1:8787/status you will see the tasks beeing completed. Once competed, gather the data: data = client . gather ( data ) df = pd . DataFrame ( data ) df . columns = [ \"number of points for MonteCarlo\" , \"value of pi\" , \"error (%)\" ] df . tail () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } number of points for MonteCarlo value of pi error (%) 1195 9697405 3.141296 0.009454 1196 9772184 3.141058 0.017008 1197 9847540 3.141616 0.000739 1198 9923477 3.141009 0.018574 1199 10000000 3.141032 0.017833 There, we have completed a simple example on how to use dask to run multiple functions in parallel. Full source code: import numpy as np import pandas as pd from distributed import Client #function used to do parallel computing on def compute_pi_MonteCarlo ( Nb_Data ): \"\"\" computes the value of pi using the monte carlo method \"\"\" Radius = 1 Nb_Data = int ( round ( Nb_Data )) x = np . random . uniform ( - Radius , Radius , Nb_Data ) y = np . random . uniform ( - Radius , Radius , Nb_Data ) pi_mc = 4 * np . sum ( np . power ( x , 2 ) + np . power ( y , 2 ) < Radius ** 2 ) / Nb_Data err = 100 * np . abs ( pi_mc - np . pi ) / np . pi return [ Nb_Data , pi_mc , err ] #connect to the scheduler client = Client ( '127.0.0.1:8786' ) #submit tasks data = [ client . submit ( compute_pi_MonteCarlo , Nb_Data ) for Nb_Data in np . logspace ( 3 , 7 , num = 1200 , dtype = int )] #gather the results data = client . gather ( data ) df = pd . DataFrame ( data ) df . columns = [ \"number of points for MonteCarlo\" , \"value of pi\" , \"error (%)\" ] df . tail () A word on the environement variables On Windows, to make sure that you can run dask-scheduler and dask-worker from the command line, you have to add the location of the executable to your path. On linux, you can append the location of the dask-worker and scheduler to the path variable with the command export PATH = $PATH :/path/to/dask","tags":"Python","url":"redoules.github.io/python/dask_distributed_parallelism.html","loc":"redoules.github.io/python/dask_distributed_parallelism.html"},{"title":"Plotting data using log axis","text":"Plotting in log axis with matplotlib import matplotlib.pyplot as plt % matplotlib inline import numpy as np x = np . linspace ( 0.1 , 20 ) y = 20 * np . exp ( - x / 10.0 ) Plotting using the standard function then specifying the axis scale One of the easiest way to plot in a log plot is to specify the plot normally and then specify which axis is to be plotted with a log scale. This can be specified by the function set_xscale or set_yscale # Normal plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . grid () plt . show () # Log x axis plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . set_xscale ( 'log' ) ax . grid () plt . show () # Log x axis plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . set_yscale ( 'log' ) ax . grid () plt . show () # Log x axis plot fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . plot ( x , y ) ax . set_xscale ( 'log' ) ax . set_yscale ( 'log' ) ax . grid () plt . show () Plotting using the matplotlib defined function Matplotlib has the function : semilogx, semilogy and loglog that can help you avoid having to specify the axis scale. # Plot using semilogx fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . semilogx ( x , y ) ax . grid () plt . show () # Plot using semilogy fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . semilogy ( x , y ) ax . grid () plt . show () # Plot using loglog fig = plt . figure () ax = fig . add_subplot ( 1 , 1 , 1 ) ax . loglog ( x , y ) ax . grid () plt . show ()","tags":"Python","url":"redoules.github.io/python/logplot.html","loc":"redoules.github.io/python/logplot.html"},{"title":"Downloading a static webpage with python","text":"If you are using python legacy (aka python 2) first of all, stop ! Furthermore, this method won't work in python legacy # Import modules from urllib.request import urlopen The webpage source code can be downloaded with the command urlopen url = \"http://example.com/\" #create a HTTP request in order to read the page page = urlopen ( url ) . read () The source code will be stored in the variable page as a string print ( page ) b '&lt;!doctype html&gt;\\n&lt;html&gt;\\n&lt;head&gt;\\n &lt;title&gt;Example Domain&lt;/title&gt;\\n\\n &lt;meta charset=\"utf-8\" /&gt;\\n &lt;meta http-equiv=\"Content-type\" content=\"text/html; charset=utf-8\" /&gt;\\n &lt;meta name=\"viewport\" content=\"width=device-width, initial-scale=1\" /&gt;\\n &lt;style type=\"text/css\"&gt;\\n body {\\n background-color: #f0f0f2;\\n margin: 0;\\n padding: 0;\\n font-family: \"Open Sans\", \"Helvetica Neue\", Helvetica, Arial, sans-serif;\\n \\n }\\n div {\\n width: 600px;\\n margin: 5em auto;\\n padding: 50px;\\n background-color: #fff;\\n border-radius: 1em;\\n }\\n a:link, a:visited {\\n color: #38488f;\\n text-decoration: none;\\n }\\n @media (max-width: 700px) {\\n body {\\n background-color: #fff;\\n }\\n div {\\n width: auto;\\n margin: 0 auto;\\n border-radius: 0;\\n padding: 1em;\\n }\\n }\\n &lt;/style&gt; \\n&lt;/head&gt;\\n\\n&lt;body&gt;\\n&lt;div&gt;\\n &lt;h1&gt;Example Domain&lt;/h1&gt;\\n &lt;p&gt;This domain is established to be used for illustrative examples in documents. You may use this\\n domain in examples without prior coordination or asking for permission.&lt;/p&gt;\\n &lt;p&gt;&lt;a href=\"http://www.iana.org/domains/example\"&gt;More information...&lt;/a&gt;&lt;/p&gt;\\n&lt;/div&gt;\\n&lt;/body&gt;\\n&lt;/html&gt;\\n' Additionally, you can beautifulsoup in order to make it easier to work with html from bs4 import BeautifulSoup soup = BeautifulSoup ( page , 'lxml' ) soup . prettify () print ( soup ) & lt ;! DOCTYPE html & gt ; & lt ; html & gt ; & lt ; head & gt ; & lt ; title & gt ; Example Domain & lt ;/ title & gt ; & lt ; meta charset = \"utf-8\" /& gt ; & lt ; meta content = \"text/html; charset=utf-8\" http-equiv = \"Content-type\" /& gt ; & lt ; meta content = \"width=device-width, initial-scale=1\" name = \"viewport\" /& gt ; & lt ; style type = \"text/css\" & gt ; body { background-color : #f0f0f2 ; margin : 0 ; padding : 0 ; font-family : \"Open Sans\" , \"Helvetica Neue\" , Helvetica , Arial , sans-serif ; } div { width : 600 px ; margin : 5 em auto ; padding : 50 px ; background-color : #fff ; border-radius : 1 em ; } a : link , a : visited { color : #38488f ; text-decoration : none ; } @ media ( max-width : 700px ) { body { background-color : #fff ; } div { width : auto ; margin : 0 auto ; border-radius : 0 ; padding : 1 em ; } } & lt ;/ style & gt ; & lt ;/ head & gt ; & lt ; body & gt ; & lt ; div & gt ; & lt ; h1 & gt ; Example Domain & lt ;/ h1 & gt ; & lt ; p & gt ; This domain is established to be used for illustrative examples in documents . You may use this domain in examples without prior coordination or asking for permission .& lt ;/ p & gt ; & lt ; p & gt ;& lt ; a href = \"http://www.iana.org/domains/example\" & gt ; More information ...& lt ;/ a & gt ;& lt ;/ p & gt ; & lt ;/ div & gt ; & lt ;/ body & gt ; & lt ;/ html & gt ;","tags":"Python","url":"redoules.github.io/python/download_page.html","loc":"redoules.github.io/python/download_page.html"},{"title":"Getting stock market data","text":"Start by importing the packages. We will need pandas and the pandas_datareader. # Import modules import pandas as pd from pandas_datareader import data Datareader allows you to import data from the internet. I have found that Quandl and robinhood works the best as a source for stockmarket data. Note that if you want an other type of data (e.g. GDP, inflation, etc.) other sources exist. #import stock from robinhood aapl_robinhood = data . DataReader ( 'AAPL' , 'robinhood' , '1980-01-01' ) aapl_robinhood . head () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } close_price high_price interpolated low_price open_price session volume symbol begins_at AAPL 2017-08-04 153.996200 154.990700 False 153.306900 153.681100 reg 20559852 2017-08-07 156.379100 156.487400 False 154.272000 154.655900 reg 21870321 2017-08-08 157.629700 159.352900 False 155.847400 156.172300 reg 36205896 2017-08-09 158.594700 158.801500 False 156.674500 156.822200 reg 26131530 2017-08-10 153.543100 158.169600 False 152.861000 158.070700 reg 40804273 #import stock from quandl aapl_quandl = data . DataReader ( 'AAPL' , 'quandl' , '1980-01-01' ) aapl_quandl . head () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } Open High Low Close Volume ExDividend SplitRatio AdjOpen AdjHigh AdjLow AdjClose AdjVolume Date 2018-03-27 173.68 175.15 166.92 168.340 38962839.0 0.0 1.0 173.68 175.15 166.92 168.340 38962839.0 2018-03-26 168.07 173.10 166.44 172.770 36272617.0 0.0 1.0 168.07 173.10 166.44 172.770 36272617.0 2018-03-23 168.39 169.92 164.94 164.940 40248954.0 0.0 1.0 168.39 169.92 164.94 164.940 40248954.0 2018-03-22 170.00 172.68 168.60 168.845 41051076.0 0.0 1.0 170.00 172.68 168.60 168.845 41051076.0 2018-03-21 175.04 175.09 171.26 171.270 35247358.0 0.0 1.0 175.04 175.09 171.26 171.270 35247358.0","tags":"Python","url":"redoules.github.io/python/stock_pandas.html","loc":"redoules.github.io/python/stock_pandas.html"},{"title":"Moving average with pandas","text":"# Import modules import pandas as pd from pandas_datareader import data , wb #import packages from pandas_datareader import data aapl = data . DataReader ( 'AAPL' , 'quandl' , '1980-01-01' ) aapl . head () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } Open High Low Close Volume ExDividend SplitRatio AdjOpen AdjHigh AdjLow AdjClose AdjVolume Date 2018-03-27 173.68 175.15 166.92 168.340 38962839.0 0.0 1.0 173.68 175.15 166.92 168.340 38962839.0 2018-03-26 168.07 173.10 166.44 172.770 36272617.0 0.0 1.0 168.07 173.10 166.44 172.770 36272617.0 2018-03-23 168.39 169.92 164.94 164.940 40248954.0 0.0 1.0 168.39 169.92 164.94 164.940 40248954.0 2018-03-22 170.00 172.68 168.60 168.845 41051076.0 0.0 1.0 170.00 172.68 168.60 168.845 41051076.0 2018-03-21 175.04 175.09 171.26 171.270 35247358.0 0.0 1.0 175.04 175.09 171.26 171.270 35247358.0 In order to computer the moving average, we will use the rolling function. #120 days moving average moving_averages = aapl [[ \"Open\" , \"High\" , \"Low\" , \"Close\" , \"Volume\" ]] . rolling ( window = 120 ) . mean () moving_averages . tail () .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } Open High Low Close Volume Date 1980-12-18 28.457667 28.551917 28.385000 28.385000 139495.000000 1980-12-17 28.410750 28.502917 28.338083 28.338083 141772.500000 1980-12-16 28.362833 28.453917 28.289167 28.289167 141256.666667 1980-12-15 28.335750 28.426833 28.262083 28.262083 144321.666667 1980-12-12 28.310750 28.402833 28.238167 28.238167 159625.000000 % matplotlib inline import matplotlib.pyplot as plt plt . plot ( aapl . index , aapl . Open , label = 'Open price' ) plt . plot ( moving_averages . index , moving_averages . Open , label = \"120 MA Open price\" ) plt . legend () plt . show ()","tags":"Python","url":"redoules.github.io/python/Moving_average_pandas.html","loc":"redoules.github.io/python/Moving_average_pandas.html"},{"title":"Keywords to use with WHERE","text":"Keywords to use with WHERE #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' Assignment operator The assignment operator is =. % sql SELECT * FROM tutyfrutty WHERE color = \"red\" * sqlite:///mydatabase.db Done. index fruit color kcal 2 Apple red 52 7 Cranberry red 308 Comparison operators Comparison operation can be done in a SQL querry. They are the following : Equality : = Greater than : > greater than or equal to : >= less than : < less than or equal to : <= not equal to : <>, != not greater than : !> not less than : !< % sql SELECT * FROM tutyfrutty WHERE kcal = 47 * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 % sql SELECT * FROM tutyfrutty WHERE kcal > 47 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE kcal >= 47 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE kcal < 47 * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 5 plum purple 28 % sql SELECT * FROM tutyfrutty WHERE kcal <= 47 * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 3 lemon yellow 15 4 lime green 30 5 plum purple 28 % sql SELECT * FROM tutyfrutty WHERE kcal <> 47 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 7 Cranberry red 308 Logical operators Logical operators test a condition and return a boolean. The logicial operators in SQL are : ALL : true if all the condtions are true AND : true is both conditions are true ANY : true if any one of the conditions are true BETWEEN : true if the operand in withing a range of values EXISTS : true if the subquery contains any rows IN : true if the condition is present in a row LIKE : true if a pattern is matched NOT : True if the operand is false, false otherwise OR : True is either condition is true SOME : true is any of the conditions is true % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" AND kcal < 100 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 3 lemon yellow 15 % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR kcal > 300 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 3 lemon yellow 15 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE fruit LIKE 'l%' * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 % sql SELECT * FROM tutyfrutty WHERE NOT color = \"yellow\" * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 2 Apple red 52 4 lime green 30 5 plum purple 28 7 Cranberry red 308 % sql SELECT * FROM tutyfrutty WHERE kcal BETWEEN 40 AND 100 * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 Bitwise operators Some bitwise operators exist in SQL. They will not be demonstrated here. They are the following : AND : & OR : | XOR : &#94; NOT : ~","tags":"SQL","url":"redoules.github.io/sql/WHERE_SQL_keywords.html","loc":"redoules.github.io/sql/WHERE_SQL_keywords.html"},{"title":"Sorting results","text":"Sorting results in SQL Sorting results can be achieved by using a modifier command at the end of the SQL querry #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' The results can be sorted with the command ORDER BY SELECT column-list FROM table_name [WHERE condition] [ORDER BY column1, column2, .. columnN] [ASC | DESC] Let's show an example where we extract the fruits that are either yellow or red % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 7 Cranberry red 308 Ascending sort % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" ORDER BY kcal ASC * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 2 Apple red 52 0 Banana yellow 89 7 Cranberry red 308 descending sort % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" ORDER BY kcal DESC * sqlite:///mydatabase.db Done. index fruit color kcal 7 Cranberry red 308 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 Sort by multiple columns You can sort by more than one column. Just specify multiple columns in the ORDER BY keyword. In the example, we will sort alphabetically on the color column first and sort alphabetically on the fruit column % sql SELECT * FROM tutyfrutty ORDER BY color , fruit ASC * sqlite:///mydatabase.db Done. index fruit color kcal 4 lime green 30 1 Orange orange 47 5 plum purple 28 2 Apple red 52 7 Cranberry red 308 0 Banana yellow 89 3 lemon yellow 15","tags":"SQL","url":"redoules.github.io/sql/Sorting_results.html","loc":"redoules.github.io/sql/Sorting_results.html"},{"title":"Filter content of a TABLE","text":"Filter content of a TABLE in SQL In this example, we will display the content of a table but we will filter out the results. Since we are working in the notebook, we will load the sql extension in order to manipulate the database. The database mydatabase.db is a SQLite database already created before the example. #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' Filter content matching exactly a condition We want to extract all the entries in a dataframe that match a certain condition, in order to do so, we will use the following command : SELECT * FROM TABLE WHERE column=\"condition\" In our example, we will filter all the entries in the tutyfrutty table whose color is yellow % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 3 lemon yellow 15 Complex conditions You can build more complex conditions by using the keywords OR and AND In the following example, we will filter all entries that are either yellow or red % sql SELECT * FROM tutyfrutty WHERE color = \"yellow\" OR color = \"red\" * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 2 Apple red 52 3 lemon yellow 15 7 Cranberry red 308 Note : when combining multiple conditions with AND and OR, be careful to use parentesis where needed Conditions matching a pattern You can also use the LIKE keyword in order to find all entries that match a certain pattern. In our example, we want to find all fruits begining with a \"l\". In order to do so, we will use the LIKE keyword and the wildcard \"%\" meaning any string % sql SELECT * FROM tutyfrutty WHERE fruit LIKE \"l%\" * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 Numerical conditions When we are working with numerical data, we can use the GREATER THAN > and SMALLER THAN < operators % sql SELECT * FROM tutyfrutty WHERE kcal < 47 * sqlite:///mydatabase.db Done. index fruit color kcal 3 lemon yellow 15 4 lime green 30 5 plum purple 28 If we want the condition to be inclusive we can use the operator <= (alternatively >=) % sql SELECT * FROM tutyfrutty WHERE kcal <= 47 * sqlite:///mydatabase.db Done. index fruit color kcal 1 Orange orange 47 3 lemon yellow 15 4 lime green 30 5 plum purple 28","tags":"SQL","url":"redoules.github.io/sql/display_table_filter.html","loc":"redoules.github.io/sql/display_table_filter.html"},{"title":"Displaying the content of a TABLE","text":"Displaying the content of a TABLE in SQL In this very simple example we will see how to display the content of a table. Since we are working in the notebook, we will load the sql extension in order to manipulate the database. The database mydatabase.db is a SQLite database already created before the example. #load the extension % load_ext sql #connect to the database % sql sqlite : /// mydatabase . db 'Connected: @mydatabase.db' In order to extract all the values from a table, we will use the following command : SELECT * FROM TABLE In our example, we want to display the data contained in the table named tutyfrutty % sql SELECT * FROM tutyfrutty * sqlite:///mydatabase.db Done. index fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 7 Cranberry red 308","tags":"SQL","url":"redoules.github.io/sql/display_table.html","loc":"redoules.github.io/sql/display_table.html"},{"title":"Opening a file with python","text":"This short article show you how to open a file using python. We will use the with keyword in order to avoid having to close the file. There is no need to import anything in order to open a file. All the function related to file manipulation are part of the python standard library In order to open a file, we will use the function open. This function takes two arguments : the path of the file the mode you want to open the file The mode can be : 'r' : read 'w' : write 'a' : append (writes at the end of the file) 'b' : binary mode 'x' : exclusive creation 't' : text mode (by default) Note that if the file does not exit it will be created if you use the following options \"w\", \"a\", \"x\". If you try to open a non existing file in read mode 'r', a FileNotFoundError will be returned. It is possible to combine multiple options together. For instance, you can open a file in binary mode for writing using the 'wb' option. Python distinguishes between binary and text I/O. Files opened in binary mode return contents as bytes objects without any decoding. In text mode , the contents of the file are returned as str, the bytes having been first decoded using a platform-dependent encoding or using the specified encoding if given. Writing to a file Let's first open (create) a text file a write a string to it. filepath = \". \\\\ myfile.txt\" with open ( filepath , 'w' ) as f : f . write ( \"Hello world !\" ) Reading a file we can now see how to read the content of a file. To do so, we will use the 'r' option with open ( filepath , \"r\" ) as f : content = f . read () print ( content ) Hello world ! A word on the with keyword In python the with keyword is used when working with unmanaged resources (like file streams). The python documentation tells us that : The with statement clarifies code that previously would use try...finally blocks to ensure that clean-up code is executed. In this section, I'll discuss the statement as it will commonly be used. In the next section, I'll examine the implementation details and show how to write objects for use with this statement. The with statement is a control-flow structure whose basic structure is: with expression [ as variable ]: with - block The expression is evaluated, and it should result in an object that supports the context management protocol (that is, has enter () and exit () methods).","tags":"Python","url":"redoules.github.io/python/Opening_file.html","loc":"redoules.github.io/python/Opening_file.html"},{"title":"Opening a SQLite database with python","text":"This short article show you how to connect to a SQLite database using python. We will use the with keyword in order to avoid having to close the database. In order to connect to the database, we will have to import sqlite3 import sqlite3 from sqlite3 import Error In python the with keyword is used when working with unmanaged resources (like file streams). The python documentation tells us that : The with statement clarifies code that previously would use try...finally blocks to ensure that clean-up code is executed. In this section, I'll discuss the statement as it will commonly be used. In the next section, I'll examine the implementation details and show how to write objects for use with this statement. The with statement is a control-flow structure whose basic structure is: with expression [ as variable ]: with - block The expression is evaluated, and it should result in an object that supports the context management protocol (that is, has enter () and exit () methods). db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : print ( \"Connected to the database\" ) #your code here except Error as e : print ( e ) Connected to the database","tags":"Python","url":"redoules.github.io/python/Opening_SQLite_database.html","loc":"redoules.github.io/python/Opening_SQLite_database.html"},{"title":"Reading data from a sql database with pandas","text":"When manipulating you data using pandas, it is sometimes useful to pull data from a database. In this tutorial, we will see how to querry a dataframe from a sqlite table. Note than it would also work with any other sql database a long as you change the connxion to the one that suits your needs. First let's import pandas and sqlite3 import pandas as pd import sqlite3 from sqlite3 import Error We want to store the table tutyfrutty in our dataframe. To do so, we will query all the elements present in the tutyfrutty TABLE with the command : SELECT * FROM tutyfrutty db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : df = pd . read_sql ( \"SELECT * FROM tutyfrutty\" , conn ) del df [ \"index\" ] #juste delete the index column that was stored in the table except Error as e : print ( e ) df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 6 Cranberry red 308 7 Cranberry red 308","tags":"Python","url":"redoules.github.io/python/Reading_data_from_a_sql_database_with_pandas.html","loc":"redoules.github.io/python/Reading_data_from_a_sql_database_with_pandas.html"},{"title":"Writing data to a sql database with pandas","text":"When manipulating you data using pandas, it is sometimes useful to store a dataframe. Pandas provides multiple ways to export dataframes. The most common consist in exporting to a csv, a pickle, to hdf or to excel. However, exporting to a sql database can prove very useful. Indeed, having a well structured database is a great for storing all the data related to your analysis in one place. In this tutorial, we will see how to store a dataframe in a new table of a sqlite dataframe. Note than it would also work with any other sql database a long as you change the connxion to the one that suits your needs. First let's import pandas and sqlite3 import pandas as pd import sqlite3 from sqlite3 import Error # Example dataframe raw_data = { 'fruit' : [ 'Banana' , 'Orange' , 'Apple' , 'lemon' , \"lime\" , \"plum\" ], 'color' : [ 'yellow' , 'orange' , 'red' , 'yellow' , \"green\" , \"purple\" ], 'kcal' : [ 89 , 47 , 52 , 15 , 30 , 28 ] } df = pd . DataFrame ( raw_data , columns = [ 'fruit' , 'color' , 'kcal' ]) df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 Now that the DataFrame has been created, let's push it to the sqlite database called mydatabase.db in a new table called tutyfrutty db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : df . to_sql ( \"tutyfrutty\" , conn ) except Error as e : print ( e ) except ValueError : print ( \"The TABLE tutyfrutty already exists, read below to understand how to handle this case\" ) Note that if the table tutyfrutty was already existing, the to_sql function will return a ValueError. This is where, the if_exists option comes into play. Let's look at the docstring of this function : \"\"\" if_exists : {'fail', 'replace', 'append'}, default 'fail' - fail: If table exists, do nothing. - replace: If table exists, drop it, recreate it, and insert data. - append: If table exists, insert data. Create if does not exist. \"\"\" Let's say, I want to update my dataframe with some new rows df . loc [ len ( df ) + 1 ] = [ 'Cranberry' , 'red' , 308 ] df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 7 Cranberry red 308 8 Cranberry red 308 I can now replace the table with the new values using the \"replace\" option db_file = \". \\\\ mydatabase.db\" try : with sqlite3 . connect ( db_file ) as conn : df . to_sql ( \"tutyfrutty\" , conn , if_exists = \"replace\" ) except Error as e : print ( e )","tags":"Python","url":"redoules.github.io/python/Writing_data_to_a_sql_database_with_pandas.html","loc":"redoules.github.io/python/Writing_data_to_a_sql_database_with_pandas.html"},{"title":"Creating a sqlite database","text":"When you want to start with using databases SQlite is a great tool. It provides an easy onramp to learn and prototype you database with a SQL compatible database. First, let's import the libraries we need import sqlite3 from sqlite3 import Error SQlite doesn't need a database server, however, you have to start by creating an empty database file import os def check_for_db_file (): if os . path . exists ( \"mydatabase.db\" ): print ( \"the database is ready\" ) else : print ( \"no database found\" ) check_for_db_file () no database found Let's then create a function that will connect to a database, print the verison of sqlite and then close the connexion to the database. def create_database ( db_file ): \"\"\" create a database connection to a SQLite database \"\"\" try : with sqlite3 . connect ( db_file ) as conn : print ( \"database created with sqlite3 version {0}\" . format ( sqlite3 . version )) except Error as e : print ( e ) create_database ( \".\\mydatabase.db\" ) database created with sqlite3 version 2.6.0 check_for_db_file () the database is ready You're all set. From now on, you can open the database and write sql querries into it.","tags":"Python","url":"redoules.github.io/python/Creating_a_sqlite_database.html","loc":"redoules.github.io/python/Creating_a_sqlite_database.html"},{"title":"Setting up the notebook for plotting with matplotlib","text":"Importing Matplotlib First we need to import pyplot, a collection of command style functions that make matplotlib work like MATLAB. Let's, as well, use the magic command %matplotlib inline in order to display the figures in the notebook import matplotlib.pyplot as plt % matplotlib inline # this doubles image size, but we'll do it manually below # %config InlineBackend.figure_format = 'retina' The following parameters are recommended for matplotlib, they will make matplotlib output a better quality image # %load snippets/matplot_setup.py plt . rcParams [ 'savefig.dpi' ] = 300 plt . rcParams [ 'figure.dpi' ] = 163 plt . rcParams [ 'figure.autolayout' ] = False plt . rcParams [ 'figure.figsize' ] = 20 , 12 plt . rcParams [ 'axes.labelsize' ] = 18 plt . rcParams [ 'axes.titlesize' ] = 20 plt . rcParams [ 'font.size' ] = 16 plt . rcParams [ 'lines.linewidth' ] = 2.0 plt . rcParams [ 'lines.markersize' ] = 8 plt . rcParams [ 'legend.fontsize' ] = 14 plt . rcParams [ 'text.usetex' ] = False # True activates latex output in fonts! plt . rcParams [ 'font.family' ] = \"serif\" plt . rcParams [ 'font.serif' ] = \"cm\" plt . rcParams [ 'text.latex.preamble' ] = \" \\\\ usepackage{subdepth}, \\\\ usepackage{type1cm}\" You can change the second line in order to fit your display. 163 dpi corresponds to a Dell Ultra HD 4k P2715Q. You can check your screen's dpi count at http://dpi.lv/","tags":"Python","url":"redoules.github.io/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html","loc":"redoules.github.io/python/Setting_up_the_notebook_for_plotting_with_matplotlib.html"},{"title":"Why using a blockchain is a bad idea for your business","text":"What having a blockchain implies? storage costs : everyone maintaining the ledger needs to store every transaction bandwith costs : everyone has to broadcast every transaction computational costs : every node has to validate the blockchain control : the creator does not control the blockchain, everyone collectively controls it developpement costs : developping on a blockchain is way harder than on a traditionnal database What to ask a business when they tell you that they are using a blockchain? When a business is telling you about their innovative technology leveraging the power of the blockchain this should immedialty spake some questions : What is the consensus algorithm? who is responsible for validating the consensus rules? what is the nature of the participation ? is it open to access? is it open to innovation? is it a public ledger? is it transparent? does it improves acountability? is it cross borders? how is it validated?","tags":"Cryptocurrencies","url":"redoules.github.io/cryptocurrencies/blockchain_bad.html","loc":"redoules.github.io/cryptocurrencies/blockchain_bad.html"},{"title":"Synology NFS share","text":"Setting up a NFS share login to your DSM admin account, open the \"Control Panel\" and go to \"File Services\" Make sure NFS is enabled Back in the control panel, go to \"Shared Folder\" Select the folder you want to share and clic \"Edit\" Go to the \"NFS Permissions tab and clic \"Create\", add the ip of the device you want to mount the mapped drive on. Make sure you copy the \"Mount path\"","tags":"Linux","url":"redoules.github.io/linux/share_nfs_share.html","loc":"redoules.github.io/linux/share_nfs_share.html"},{"title":"Mount a NFS share using fstab","text":"Mount nfs using fstab The fstab file, generally located at /etc/fstab lists the differents partitions and where to load them on the filesystem. You can edit this file as root by using the following command sudo nano /etc/fstab in the following example, we want to mount a NFS v3 share from : server : 192.168.1.2 mountpoint (on the server) : /volumeUSB2/usbshare * mountlocation (on the client) : /mnt we specify 192 .168.1.2:/volumeUSB2/usbshare /mnt nfs nfsvers = 3 ,users 0 0 the client will then automatically mount the share ont /mnt at startup. Related you can reload the fstab file using this method : https://redoules.github.io/linux/Reloading_fstab.html You can create a NFS share on a Synology using the method : https://redoules.github.io/linux/share_nfs_share.html","tags":"Linux","url":"redoules.github.io/linux/mount_nfs_share_fstab.html","loc":"redoules.github.io/linux/mount_nfs_share_fstab.html"},{"title":"Installing bitcoind on raspberry pi","text":"Installing bitcoind on linux Running a full bitcoin node helps the bitcoin network to accept, validate and relay transactions. If you want to volunteer some spare computing and bandwidth resources to run a full node and allow Bitcoin to continue to grow you can grab an inexpensive and power efficient raspberry pi and turn it into a full node. There are plenty of tutorials on the Internet explaining how to install a bitcoin full node; this tutorial won't go over setting up a raspberry pi and using ssh. In order to store the full blockchain we will mount a network drive and tell bitcoind to use this mapped drive as the data directory. Download the bitcoin client Go to https://bitcoin.org/en/download Copy the URL for the ARM 32 bit version and download it onto your raspberry pi. wget https://bitcoin.org/bin/bitcoin-core-0.15.1/bitcoin-0.15.1-arm-linux-gnueabihf.tar.gz Locate the downloaded file and extract it using the arguement xzf tar xzf bitcoin-0.15.1-arm-linux-gnueabihf.tar.gz a new directory bitcoin-0.15.1 will be created, it contrains the files we need to install the software Install the bitcoin client We will install the content by copying the binaries located in the bin folder into /usr/local/bin by using the install command. You must use sudo because it will write data to a system directory sudo install -m 0755 -o root -g root -t /usr/local/bin bitcoin-0.15.1/bin/* Launch the bitcoin core client by running bitcoind -daemon Configuration of the node Start your node at boot Starting you node automatically at boot time is a good idea because it doesn't require a manual action from the user. The simplest way to achive this is to create a cronjob. Run the following command crontab -e Select the text editor of your choice, then add the following line at the end of the file @reboot bitcoind -daemon Save the file and exit; the updated crontab file will be installed for you. Full Node If you can afford to download and store all the blockchain, you can run a full node. At the time of writing, the blockchain is 150Go ( https://blockchain.info/fr/charts/blocks-size ). Tree ways to store this are : use a microSD with 256Go or more add a thumbdrive or an external drive to your raspberry pi * mount a network drive from a NAS If you have purchased a big SD card then you can leave the default location for the blockchain data (~/.bitcoin/). Otherwise, you will have to change the datadir location to where your drive is mounted (in my case I have mounted it to /mnt) In order to configure your bitcoin client, edit/create the file bitcoin.conf located in ~/.bitcoin/ nano ~/.bitcoin/bitcoin.conf copy the following text # From redoules.github.io # This config should be placed in following path: # ~/.bitcoin/bitcoin.conf # [core] # Specify a non-default location to store blockchain and other data. datadir=/mnt # Set database cache size in megabytes; machines sync faster with a larger cache. Recommend setting as high as possible based upon mach$ dbcache=100 # Keep at most <n> unconnectable transactions in memory. maxorphantx=10 # Keep the transaction memory pool below <n> megabytess. maxmempool=50 # [network] # Maintain at most N connections to peers. maxconnections=40 # Tries to keep outbound traffic under the given target (in MiB per 24h), 0 = no limit. maxuploadtarget=5000 Check https://jlopp.github.io/bitcoin-core-config-generator it is a handy site to edit the bitcoin.conf file Pruning node If you don't want to store the entire blockchain you can run a pruning node which reduces storage requirements by enabling pruning (deleting) of old blocks. Let's say you want to allocated at most 5Go to the blockchain, then specify prune=5000 into your bitcoin.conf file. Edit/create the file bitcoin.conf located in ~/.bitcoin/ nano ~/.bitcoin/bitcoin.conf copy the following text # From redoules.github.io # This config should be placed in following path: # ~/.bitcoin/bitcoin.conf # [core] # Set database cache size in megabytes; machines sync faster with a larger cache. Recommend setting as high as possible based upon mach$ dbcache=100 # Keep at most <n> unconnectable transactions in memory. maxorphantx=10 # Keep the transaction memory pool below <n> megabytess. maxmempool=50 # Reduce storage requirements by only storing most recent N MiB of block. This mode is incompatible with -txindex and -rescan. WARNING: Reverting this setting requires re-downloading the entire blockchain. (default: 0 = disable pruning blocks, 1 = allow manual pruning via RPC, greater than 550 = automatically prune blocks to stay under target size in MiB). prune=5000 # [network] # Maintain at most N connections to peers. maxconnections=40 # Tries to keep outbound traffic under the given target (in MiB per 24h), 0 = no limit. maxuploadtarget=5000 Checking if your node is public one of the best way to help the bitcoin network is to allow your node to be visible and to propagate block to other nodes. The bitcoin protocole uses port 8333, other clients should be able to share information with your client. Run ifconfig and check if you have an ipv6 adresse (look for adr inet6:) IPV6 Get the global ipv6 adresse of your raspberry pi Link encap:Ethernet HWaddr xx:xx:xx:xx:xx:xx inet adr:192.168.1.x Bcast:192.168.1.255 Masque:255.255.255.0 adr inet6: xxxx::xxxx:xxxx:xxxx:xxxx/64 Scope:Lien adr inet6: xxxx:xxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx/64 Scope:Global UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:42681744 errors:0 dropped:0 overruns:0 frame:0 TX packets:38447218 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 lg file transmission:1000 RX bytes:3044414780 (2.8 GiB) TX bytes:2599878680 (2.4 GiB) it is located between adr inet4 and Scope:Global adr inet6: xxxx:xxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx/64 Scope:Global Copy this adresse and past it into the search field on https://bitnodes.earn.com/ If your node is visible, it will appear on the website IPV4 If you don't have an ipv6 adresse, you will have to open port 8333 on your router and redirect it to the internal IP of your raspberry pi. It is not detailed here because the configuration depends on your router.","tags":"Cryptocurrencies","url":"redoules.github.io/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html","loc":"redoules.github.io/cryptocurrencies/Installing_bitcoind_on_raspberry_pi.html"},{"title":"Reloading .bashrc","text":"Reload .bashrc The .bashrc file, located at ~/.bashrc allows a user to personalize its bash shell. If you edit this file, the changes won't be loaded without login out and back in. However, you can use the following command to do it source ~/.bashrc","tags":"Linux","url":"redoules.github.io/linux/Reloading_.bashrc.html","loc":"redoules.github.io/linux/Reloading_.bashrc.html"},{"title":"Reloading fstab","text":"Reload fstab The fstab file, generally located at /etc/fstab lists the differents partitions and where to load them on the filesystem. If you edit this file, the changes won't be automounted. You either have to reboot your system of use the following command as root mount -a","tags":"Linux","url":"redoules.github.io/linux/Reloading_fstab.html","loc":"redoules.github.io/linux/Reloading_fstab.html"},{"title":"Updating all python package with anaconda","text":"Updating anaconda packages All packages managed by conda can be updated with the following command : conda update --all Updating other packages with pip For the other packages, the pip package manager can be used. Unfortunately pip hasn't the same update all fonctionnality. import pip from subprocess import call for dist in pip . get_installed_distributions (): print ( \"updating {0}\" . format ( dist )) call ( \"pip install --upgrade \" + dist . project_name , shell = True )","tags":"Python","url":"redoules.github.io/python/updating_all_python_package_with_anaconda.html","loc":"redoules.github.io/python/updating_all_python_package_with_anaconda.html"},{"title":"Saving a matplotlib figure with a high resolution","text":"creating a matplotlib figure #Importing matplotlib % matplotlib inline import matplotlib.pyplot as plt import numpy as np Drawing a figure # Fixing random state for reproducibility np . random . seed ( 19680801 ) mu , sigma = 100 , 15 x = mu + sigma * np . random . randn ( 10000 ) # the histogram of the data n , bins , patches = plt . hist ( x , 50 , normed = 1 , facecolor = 'g' , alpha = 0.75 ) plt . xlabel ( 'Smarts' ) plt . ylabel ( 'Probability' ) plt . title ( 'Histogram of IQ' ) plt . text ( 60 , . 025 , r '$\\mu=100,\\ \\sigma=15$' ) plt . axis ([ 40 , 160 , 0 , 0.03 ]) plt . grid ( True ) plt . show () Saving the figure normally, one would use the following code plt . savefig ( 'filename.png' ) &lt;matplotlib.figure.Figure at 0x2e45e92f400&gt; The figure in then exported to the file \"filename.png\" with a standard resolution. In adittion, you can specify the dpi arg to some scalar value, for example: plt . savefig ( 'filename_hi_dpi.png' , dpi = 300 ) &lt;matplotlib.figure.Figure at 0x2e462164898&gt;","tags":"Python","url":"redoules.github.io/python/Saving_a_matplotlib_figure_with_a_high_resolution.html","loc":"redoules.github.io/python/Saving_a_matplotlib_figure_with_a_high_resolution.html"},{"title":"Iterating over a DataFrame","text":"Create a sample dataframe # Import modules import pandas as pd # Example dataframe raw_data = { 'fruit' : [ 'Banana' , 'Orange' , 'Apple' , 'lemon' , \"lime\" , \"plum\" ], 'color' : [ 'yellow' , 'orange' , 'red' , 'yellow' , \"green\" , \"purple\" ], 'kcal' : [ 89 , 47 , 52 , 15 , 30 , 28 ] } df = pd . DataFrame ( raw_data , columns = [ 'fruit' , 'color' , 'kcal' ]) df .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } fruit color kcal 0 Banana yellow 89 1 Orange orange 47 2 Apple red 52 3 lemon yellow 15 4 lime green 30 5 plum purple 28 Using the iterrows method Pandas DataFrames can return a generator with the iterrrows method. It can then be used to loop over the rows of the DataFrame for index , row in df . iterrows (): print ( \"At line {0} there is a {1} which is {2} and contains {3} kcal\" . format ( index , row [ \"fruit\" ], row [ \"color\" ], row [ \"kcal\" ])) At line 0 there is a Banana which is yellow and contains 89 kcal At line 1 there is a Orange which is orange and contains 47 kcal At line 2 there is a Apple which is red and contains 52 kcal At line 3 there is a lemon which is yellow and contains 15 kcal At line 4 there is a lime which is green and contains 30 kcal At line 5 there is a plum which is purple and contains 28 kcal","tags":"Python","url":"redoules.github.io/python/Iterating_over_a_dataframe.html","loc":"redoules.github.io/python/Iterating_over_a_dataframe.html"},{"title":"Article Recommander","text":"import pandas as pd import numpy as np % matplotlib inline Loading data and preprocessing we first learn the pickled article database. We will be cleaning it and separating the interesting articles from the uninteresting ones. df = pd . read_pickle ( './article.pkl' ) del df [ \"html\" ] del df [ \"image\" ] del df [ \"URL\" ] del df [ \"hash\" ] del df [ \"source\" ] df [ \"label\" ] = df [ \"note\" ] . apply ( lambda x : 0 if x <= 0 else 1 ) df . head ( 5 ) .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } authors note resume texte titre label 0 [Danny Bradbury, Marco Santori, Adam Draper, M... -10.0 Black Market Reloaded, a black market site tha... Black Market Reloaded, a black market site tha... Black Market Reloaded back online after source... 0 1 [Emily Spaven, Stan Higgins, Emilyspaven] 1.0 The UK Home Office believes the government sho... The UK Home Office believes the government sho... Home Office: UK Should Create a Crime-Fighting... 1 2 [Pete Rizzo, Alex Batlin, Yessi Bello Perez, P... -10.0 Though lofty in its ideals, lead developer Dan... A new social messaging app is aiming to disrup... Gems Bitcoin App Lets Users Earn Money From So... 0 3 [Nermin Hajdarbegovic, Stan Higgins, Pete Rizz... 3.0 US satellite service provider DISH Network has... US satellite service provider DISH Network has... DISH Becomes World's Largest Company to Accept... 1 4 [Stan Higgins, Bailey Reutzel, Garrett Keirns,... -10.0 An unidentified 28-year-old man was robbed of ... An unidentified 28-year-old man was robbed of ... Bitcoin Stolen at Gunpoint in New York City Ro... 0 Basic statistics on the dataset let's explore the dataset and extract some numbers : * the number of article liked/disliked df [ \"label\" ] . value_counts () 0 879 1 324 Name: label, dtype: int64 Create the full content column df [ 'full_content' ] = df . titre + ' ' + df . resume #exclude the full texte of the article for the moment df . head ( 1 ) .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } authors note resume texte titre label full_content 0 [Danny Bradbury, Marco Santori, Adam Draper, M... -10.0 Black Market Reloaded, a black market site tha... Black Market Reloaded, a black market site tha... Black Market Reloaded back online after source... 0 Black Market Reloaded back online after source... from sklearn.model_selection import train_test_split training , testing = train_test_split ( df , # The dataset we want to split train_size = 0.75 , # The proportional size of our training set stratify = df . label , # The labels are used for stratification random_state = 400 # Use the same random state for reproducibility ) training . head ( 5 ) .dataframe thead tr:only-child th { text-align: right; } .dataframe thead th { text-align: left; } .dataframe tbody tr th { vertical-align: top; } authors note resume texte titre label full_content 748 [Jon Brodkin] -10.0 Amazon, Reddit, Mozilla, and other Internet co... Amazon, Reddit, Mozilla, and other Internet co... Amazon and Reddit try to save net neutrality r... 0 Amazon and Reddit try to save net neutrality r... 1183 [Jon Brodkin] -10.0 (The Time Warner involved in this transaction ... A group of mostly Democratic senators led by A... Democrats urge Trump administration to block A... 0 Democrats urge Trump administration to block A... 769 [Joseph Brogan] -10.0 On Twitter, bad news comes at all hours, with ... On Twitter, bad news comes at all hours, with ... Some of the best art on Twitter comes from the... 0 Some of the best art on Twitter comes from the... 57 [Michael Del Castillo, Pete Rizzo, Trond Vidar... -10.0 Publicly traded online travel service Webjet i... Publicly traded online travel service Webjet i... Webjet Ethereum Pilot Targets Hotel Industry's... 0 Webjet Ethereum Pilot Targets Hotel Industry's... 892 [Andrew Cunningham] 10.0 What has changed on the 2017 MacBook, then?\\nI... Andrew Cunningham\\n\\nAndrew Cunningham\\n\\nAndr... Mini-review: The 2017 MacBook could actually b... 1 Mini-review: The 2017 MacBook could actually b... from sklearn.feature_extraction.text import TfidfVectorizer , CountVectorizer from sklearn.svm import LinearSVC , SVC from sklearn.pipeline import Pipeline from sklearn.model_selection import cross_val_predict from utils.plotting import pipeline_performance steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , LinearSVC ()) ) pipeline = Pipeline ( steps ) predicted_labels = cross_val_predict ( pipeline , training . full_content , training . label ) pipeline_performance ( training . label , predicted_labels ) pipeline = pipeline . fit ( training . titre , training . label ) Accuracy = 80.6% Confusion matrix, without normalization [[624 35] [140 103]] import re from utils.plotting import print_top_features from sklearn.model_selection import GridSearchCV def mask_integers ( s ): return re . sub ( r '\\d+' , 'INTMASK' , s ) steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , LinearSVC ()) ) pipeline = Pipeline ( steps ) gs_params = { #'vectorizer__use_idf': (True, False), 'vectorizer__lowercase' : [ True , False ], 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__C' : np . linspace ( 5 , 20 , 25 ) } gs = GridSearchCV ( pipeline , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline1 = gs . best_estimator_ predicted_labels = pipeline1 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline1 , n_features = 10 ) aaa = gs . predict ( testing . full_content ) == testing . label aaa = aaa [ testing . label == 1 ] testing [ \"titre\" ] . iloc [ ~ aaa . values ] #pipeline1.predict([\"windows xbox bitcoin\"]) from sklearn.externals import joblib joblib . dump ( pipeline1 , 'classifier.pkl' ) gs . predict ([ 'Google' ]) array([1], dtype=int64) steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , SVC ()) ) pipeline = Pipeline ( steps ) gs_params = { #'vectorizer__use_idf': (True, False), 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__C' : np . linspace ( 5 , 20 , 25 ) } gs = GridSearchCV ( pipeline , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline1 = gs . best_estimator_ predicted_labels = pipeline1 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline1 , n_features = 10 ) {'classifier__C': 5.0, 'vectorizer__ngram_range': (1, 1), 'vectorizer__preprocessor': &lt;function mask_integers at 0x00000237491B67B8&gt;, 'vectorizer__stop_words': 'english'} 0.711180124224 Accuracy = 71.2% Confusion matrix, without normalization [[153 0] [ 62 0]] --------------------------------------------------------------------------- ValueError Traceback (most recent call last) &lt;ipython-input-9-3e0781e307fb&gt; in &lt;module&gt;() 25 pipeline_performance(testing.label, predicted_labels) 26 ---&gt; 27 print_top_features(pipeline1, n_features=10) C:\\Users\\Guillaume\\Documents\\Code\\recommandation\\utils\\plotting.py in print_top_features(pipeline, vectorizer_name, classifier_name, n_features) 81 def print_top_features(pipeline, vectorizer_name='vectorizer', classifier_name='classifier', n_features=7): 82 vocabulary = np.array(pipeline.named_steps[vectorizer_name].get_feature_names()) ---&gt; 83 coefs = pipeline.named_steps[classifier_name].coef_[0] 84 top_feature_idx = np.argsort(coefs) 85 top_features = vocabulary[top_feature_idx] C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\svm\\base.py in coef_(self) 483 def coef_(self): 484 if self.kernel != 'linear': --&gt; 485 raise ValueError('coef_ is only available when using a ' 486 'linear kernel') 487 ValueError: coef_ is only available when using a linear kernel from sklearn.naive_bayes import BernoulliNB steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , BernoulliNB ()) ) pipeline2 = Pipeline ( steps ) gs_params = { 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__alpha' : np . linspace ( 0 , 1 , 5 ), 'classifier__fit_prior' : [ True , False ] } gs = GridSearchCV ( pipeline2 , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline2 = gs . best_estimator_ predicted_labels = pipeline2 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline2 , n_features = 10 ) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:820: RuntimeWarning: divide by zero encountered in log neg_prob = np.log(1 - np.exp(self.feature_log_prob_)) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:823: RuntimeWarning: invalid value encountered in add jll += self.class_log_prior_ + neg_prob.sum(axis=1) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:801: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - {'classifier__alpha': 0.25, 'classifier__fit_prior': True, 'vectorizer__ngram_range': (1, 1), 'vectorizer__preprocessor': &lt;function mask_integers at 0x00000237491B67B8&gt;, 'vectorizer__stop_words': 'english'} 0.805900621118 Accuracy = 78.1% Confusion matrix, without normalization [[140 13] [ 34 28]] Top like features: ['use' 'just' 'year' 'price' 'time' 'Bitcoin' 'bitcoin' 'new' 'The' 'INTMASK'] --- Top dislike features: ['ABBA' 'cable' 'cab' 'byte' 'publication' 'bye' 'publications' 'publicity' 'buyer' 'publicizing'] from sklearn.naive_bayes import MultinomialNB steps = ( ( 'vectorizer' , TfidfVectorizer ()), ( 'classifier' , MultinomialNB ()) ) pipeline3 = Pipeline ( steps ) gs_params = { 'vectorizer__stop_words' : [ 'english' , None ], 'vectorizer__ngram_range' : [( 1 , 1 ), ( 1 , 2 ), ( 2 , 2 )], 'vectorizer__preprocessor' : [ mask_integers , None ], 'classifier__alpha' : np . linspace ( 0 , 1 , 5 ), 'classifier__fit_prior' : [ True , False ] } gs = GridSearchCV ( pipeline3 , gs_params , n_jobs = 1 ) gs . fit ( training . full_content , training . label ) print ( gs . best_params_ ) print ( gs . best_score_ ) pipeline3 = gs . best_estimator_ predicted_labels = pipeline3 . predict ( testing . full_content ) pipeline_performance ( testing . label , predicted_labels ) print_top_features ( pipeline3 , n_features = 10 ) C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - C:\\Users\\Guillaume\\Anaconda3\\lib\\site-packages\\sklearn\\naive_bayes.py:699: RuntimeWarning: divide by zero encountered in log self.feature_log_prob_ = (np.log(smoothed_fc) - {'classifier__alpha': 0.5, 'classifier__fit_prior': False, 'vectorizer__ngram_range': (1, 1), 'vectorizer__preprocessor': &lt;function mask_integers at 0x00000237491B67B8&gt;, 'vectorizer__stop_words': 'english'} 0.80900621118 Accuracy = 79.1% Confusion matrix, without normalization [[141 12] [ 33 29]] Top like features: ['time' 'Google' 'Pro' 'Apple' 'new' 'The' 'Bitcoin' 'price' 'bitcoin' 'INTMASK'] --- Top dislike features: ['ABBA' 'categories' 'catching' 'catalyst' 'catalog' 'casually' 'casts' 'cast' 'cashier' 'ran']","tags":"Machine Learning","url":"redoules.github.io/machine-learning/Source code for the recommandation engine for articles.html","loc":"redoules.github.io/machine-learning/Source code for the recommandation engine for articles.html"}]}
\ No newline at end of file