From 47996892cdfedfa816add1ea4c9566c99b4e4cca Mon Sep 17 00:00:00 2001
From: Katarina Elez <15355014+katarinaelez@users.noreply.github.com>
Date: Thu, 24 Jan 2019 22:13:51 +0100
Subject: [PATCH] First commit

---
 models/model.npz   | Bin 0 -> 12887 bytes
 src/gor-predict.py |  44 ++++++++++++++++++++++++++++++++++++++
 src/gor-train.py   |  49 +++++++++++++++++++++++++++++++++++++++++++
 src/svm-predict.py |  51 +++++++++++++++++++++++++++++++++++++++++++++
 src/svm-train.py   |  51 +++++++++++++++++++++++++++++++++++++++++++++
 src/utils.py       |  39 ++++++++++++++++++++++++++++++++++
 6 files changed, 234 insertions(+)
 create mode 100644 models/model.npz
 create mode 100644 src/gor-predict.py
 create mode 100644 src/gor-train.py
 create mode 100644 src/svm-predict.py
 create mode 100644 src/svm-train.py
 create mode 100644 src/utils.py

diff --git a/models/model.npz b/models/model.npz
new file mode 100644
index 0000000000000000000000000000000000000000..13270fa4dcc4cf42a1b5605efc35d45f7c50de6e
GIT binary patch
literal 12887
zcmbuGXH!;P^W_1>047ufL`75t3@8{dattUa1{6@tB7y+~FGO!KAVIR^AV?O;IY<zM
zs_ri`Rr5V&=C`@$`A^M@c{N%pT%5CaukO{WyZ1RpMl4yn(eB@0-gf`3&voU0545v0
ze^&D0tJJ3n$pI;^Gpet}T#dYAZ}-mbqwmv%H}Pq{C${^3y6(GgyKiD@T6$Vs%A?e@
zrwQici*d<s68QYh^SIXueDCMKZ~N!%|9`)nvK-=5-oARB5fGo6mJsmjZF2gHxU{sm
z45vqF3Gt~ZZ_?A=#-}^k?EbyVDdp+EKegFM*yY(A;vdBWB>s(4LW=pq$+jfc9&}G^
zOJkSuZ*pAvi<Cf{W4>+KP22JU+lrgEm3|TS{{MTir|B85xyLEiforcHvpGlDMc5bE
zR^7C%_KWy;rBAjs*REZ={6G1%_*aCLZd?1L=z(qBXPb-P;%4Et_0R2|FWR%=Ih$;}
zX>(<h2$zVB`L<2Z|6M)tn9c3q)!lE}JpQx#Dck1%dv&{Xo9C0F=SyN2*4o)=m3-5M
zWzup~CN24SUFse$%aA@xyL3wHr6gQu<(DOcS`jaKQu9uRto1skWnr4HQ&MC}<vz*M
zHtCjHt~=s&>w(U2-<b76a$oDPPRN3cSX!!0zNryXDc>!x1Q*S*mo6E!WXj!t<wc#B
zX-gYr_`EdfkBjW!d48D;I_nHqORfy*H<^$!`5_%vjO5;zAsM?OU6yo*U6kQ3!L0ux
zdZRRG7kl+vk}XB&FKGEaDYQH@BBf7zoTSY<oGjVbwO9UfxqOppnR>w5qoCFion98M
z9nt{qTiWtPMl-YzG<Of{qzpMr6DGYnl<us#ItJu0k0T)tYT)h(kF&oPBI8z&RA*{d
zcr0sgJ*Q>*Q~qqh_;WJvcS<@W2dvtvh1axslc&sBE?x%1iKDEmlb^v~Wm_fnmJX!r
zOv+yT++l^u=(=Fct&~m+xBS#W&61fwHb^=qC7a(%A5eAf%KP-l2!Kkj75@3S)=8@s
zpjmgM7=QJ1aX_?8d=3qiDeJs`*MW>(Qe{BO?8^jqX$8W$y*E#>VFRY@)Ov#2AO)5b
zTaV8L=;-Pl?0&=dUg+^JfQ;22Z;#z$fniJ99PcJbtJX-JCBv`QN}GPwW>7{&NxjTy
zK5Uq=>?G^TN4G$hJ|$mcox{!>j0-+m<P#|6V0;_zCEt9t0yBa%N5A3ppN=Lx+3)=y
zOS|NoXx2Y`FLg3;4UliOz(oq)90sI}OW9$qInF*K*6qaS`vNr=kdD&J?`+N1VPHB+
zTPS|YeZU4<t6359Q)*H%=cMMZfYK#|x5$cFE^}JEjv#ef$M5T;{Wcu}?Gc^0<0bR4
zTps#3^(KDF0qo`@(s@QkJo!|bR-b;b)kXeTtIj->nWvIvJ{<N)k6)Q$fL8312{2WL
zgVfXY;FrTX&R18pal4eTS_y9SmO`nBma3fqb&{6w;9bGM4AT}Jlp^^HC0nd8ZykN_
z`o_4WPP1ivy=H}yFI@SH4{|MTwUf@6F!>t@Odsup+S%H831hV0QO2EgdIu<@w2-I_
z?AT-yXJqe&WgcH-!dglM2b_ZbkpXM-Ass%w`6w85>bA5miFwTm&PnGSrQ{13Bc<)g
z`DpFu!`J%PO?rX|P`wq7@5+vA_9me1*9xr<ldsw%KdmEgu;Padcp0`#OYZVC`L2Zo
zg?}ozVoVEcVAa3ocZlR^9$2MAC$#UfHs1&CrKK_$DrM&(lIEXUi<V@+2W*D)YWIOv
zQp%1xfA_prAJ$Ts=aH!Shh^9StoxH5hDcGYK{_9P_QC@#04j`|OG2jl12Dot$G>Pr
zrZ!0}-U!n1c<tSfrNO7=_eK5oQD&_WEqSVy9xM3tA_4iT3u|{tgSAHU&PwCs7wj$z
zE=S@dJ6_tYw0-UfOWPU4fS<_U^gH?P#XkB~^1gUPxnk5l2g!fustdpba{HPiGI1M}
zPFjVOe6wA~$2#W{@Ca7+15;<Moyb|wQa0L<B%@j^J$}HHqWBaUdImuXtrr*k9AltC
zmxV)HR%z>PSNUzFEY~uVb=}tWO{<Ce;0gnB(mNCkzsZteYwc&E)}htT^4F3-tE6y;
zwk7g;=o{Co_E59I!ZE|zx)k;4X8+8`moGW(A?uv9EllfIA)kYm%NG50MgI7pJle=M
zMN$`Y1mU<C|3Jod+>Of}mTRk{Ozbz@ZP1ESAHvt*SoE+@<^wkEUPSYZ`ZB6zG5}|y
zq&H3qo*~Nv)<K<LujB9EFLK0xS?aFa0Q#i0DaX<51z<vgQ`#0`PuA*GnEsZMV2Vu*
z-)NJA{9b~X3|p>9+Ba7F2c%(#(2LNw7_-vHPW#q`fb+DDEM5POT<fs>JWK^~rqcYd
zq>)wrQb@{tF*2G=9Fma$7}R4)yAG#Q1(CU83$ABD?t$2qi;|QGJRQZ9VQb4yEk|8S
zuE?BqdL6<0;hShO^4CquZTWI`X*0W2?zwvUDy#Zz47sTb34oOWZF?j?qi*tP`ZHe{
zy71Bz<}$6bKkyXM_(|DQLM}beoQP|zW_cp)1<3t~m8$c0?lGTa0oC-<Nu8FqD>91j
zFGaZDmmYxa@_K*}w#rwdh+lQ|^|C0cX04^YT6oigD&0vqS8Ba~+OB(sRx_KWW`%Z~
zmJ&<ycIwb(C7Nwks&~qp;~&_Og7bI#F=@B}OlkgPm%s;=;PgamS|s=7GwjXuNZuw@
zhrx5&=d9L!G}>lK$r3xsIfJ0#(iX@4=Vg*|PgJ{>KS*&gsjIzr^oOr}TW=I8PYRN$
zG5LGg^VXinK35YB$QdU;e{GBd;-zo|<rmWO2aw*Y&g|FROF;C}9xYlfZB!D1RTs1G
zLIP&>SwY@1_Cd=xm~HB|QDs~2l9)vnbz_cvoYfp3P%hUNdXFk;SkA8dwHkphTt!u#
zwV?aN)%%87V11#(FViH)0}dUK5gmOa1uTa4)S`C-I`BT3&Eu%wvvI4Q0(YHu+`Hk8
zwvSuVnjY~2a!eUr9F?A9(sSYlc<qlk+oUs`ZI{F!H3{?sxFkxIWM9*w4*<^8@|Ogy
z%1s(+!KBj-Mh!#*`IY?KkDzq=NV&B%`RWqeYu4U%djAbAScg&Dq*-&5bm+N>!4IvJ
zs_k!BxYhb_LuOFcT1+2is~9Y=u-^dIP-@uB)-*JB$Xd4)wdo{3e*!kdTSgtFdM~T)
z3D!BvqY;5(OZ&BwIMi$ZRZ><9kH~^M-`CO;IzUNkxs2Z;Cqo`<vv!{T1YT+W>>i?H
zpADmGu1nEJ7hD437Oa@VssWh;Gm+`nF8`&@26L$lySW6Ag*2xNQkq0jY1kpv^cN2U
z@pUJ1+-C(odd=6rhyzVjP5cT-GR^YV8!bTpW~9i&q+y8;Z``+rY#p||Jhday)F6t^
zUL6mQiU<O>bRCpGu{J3-SyL6k&%f{VC6aEeH>VS+-iBR`I_ECcp$EY1nz~d<j-N0A
z{*^9+2|n9d1ly-DWZcV@#_-GQdu!I6V8fHEH1B}6y2v~&jTUKV+`*+dN@`$=wrUp5
zQG>M$&(5;zEb*e~&8~~T^Ej5zrSOvcmZHrR912GJE^T%?9%DFNq+^$4ENFw#lWIFR
zTHx=Hm#(Ro6dBMV9~m*Qx>>hYy4Gw2t>i`Pg2Ve`=pBvRgpRKDBGTX;xN<;>aD6R+
zH|V_Py_cFxCd@yz`si~le8Ywpx3K*8M0Ob?;4-#%ss2eZJo^43LEDLmiE-8S=ui6l
zI3M4J20!KJrg&6&b>tn|+a@xJ02h(tKl~y{$a?FR=d#N(;$TduplOiu_+6$$m^lk;
zN~H5P4BB&7s&}G(J(xf!I<#OXWc_Ix*f9t+i*xE|O|8?(=eNqAXm;ka;?P^`{b-=%
zx4SJ&^ER;#_-gEzorjuiR-m05$?w{2DB<R_Nfd=yHzR*Jm^_<soyHR`jqq{zW2w1o
z01xYIivA9FG3>3QQe-nH<ocM|r8vs*lXk4*2T#oeM26)*0Q1K%{jo}BR~pFeTC4*b
z5^2Eet=FHup6!h^M)U=d)FqUluf}}ox`6ETQ#c2#G%8MwPI{e%XCIQa_I$ugQzCOt
zqFq^%eSqsJL1o%RReNMC>WXlUz7$Pw9P$3WHaP1DTuHnpo$n68@?X{tGP+C_H0vn&
zk|BB9Lp6((u$&jwVVHrS2f#6rVZ^n9S_e0KR$oYywtrY$W?pM|i1hmdR<pd7@ok6X
z89z_Sl#BkfX*UY8@)Dv`6B7j7z%UfK_;rH0zV<5K>WyRF4lB)GM^3vtQZ0NBXz|@G
zk%orVLHg&^E3E7`g*H#iGjF94Tdg(;D9w>N2BqmeCeocDcDJ#oQL+=IED+J2v~Eag
zY9@8)4}kX{*tBEMK7hPPTuyX4!uc%Vo(qVIwcKQH@1}M7>mU;?-9*DNGOZ0CPJs5f
z7P~uUk_xOoy-Au}B|FJj-z25vW?k4igLe2I{k@!lqJSojX|esP<Q>=zTx4r-XRtTy
zt+PU<npN^PLZl9+%<9ziFogG~^$A(2Md>=J1A|CPdpZI+O89%|7If&qO|bSG*wdd+
z?u_TlR_mF}9zGZh(MK%XmggKiY`tmQmTvQUvUpT*!M6QBM+M&Is9?uU+s^+wD%kbh
z!%Y7F!%Y4KGAtE-sdvz+d(pQxtfx-)n)bDmE~Zu6-Hm#WBs~q>d<u+v{387%hxV}?
zwe~uw4VlZ078RKyUY16@@lS{RTCT-$6ygaE35N9R)%za!O4={`=-=4$uy<+fdJnE`
zFqOXNwy!PP-MFt7k^g;PrU)AXUZutr6|bd-4W#S1*8Za<w46QG3#ZpeLl$Gjh~+D7
zP65mJ8J@KJZrdt-_`Sr^KUcLr`V|~Rb$%_A=BrDI(x~;E>HO&}JUM2ib9?E92cN)t
zOvjQO0GnrgJ@6<NGK?K%?N3)^o&_su_{uG3{d#n(RIM>+zaKtLNUP}~KwE;PHLFgM
zKO7-cUI}^PMhOA!6zyWRl=J5+DRqS@^@6?B;F1K_%cR;L9R_oXHYVxVsp9~4lrG;$
zRG<hl=pU%pk1YROg?%4(C$E?Bb=bd>!u(^0b_Re~a_r1(Rf=^Gw>?_>v~evZvd!{L
zzoP}Ky}15?R^66A$90siwQ>ilI4`xSEJ+QTysp%ZLGw+6cD}uAuQ|Y3b6AD~9L|{V
zemn5``m)u=`Cnfy-wR2q`1xfmdzeh?R%4`dl7QrSYStq}EA9=XV3<H!do6$I|0h-}
zP)X~!zXK8d^Nf4L$;;|p&<3<JyPu4t<f1F>ly1*07ZLq$+$#gS&Ui?RFDA=)0`+_@
zC;_-v&PK6cE<Jjob?75^b<5lZqjDo2(56<Za2zuC^^w*u`;g4d2fd_A2kvge`BE3E
z12=Xqn#Np{i}h{Nidc+T2_Z|R?pmUhlGZzL;V%k^iGYv)3cIJ~9A_J#hISL~0jlHK
zRmZjeJzLjX*V34O>GSF_`DBn*u9S((rV7++?eYLz*=4L5KbQENR=8cNEX{TLazk1&
z3?qwW=HMn~$3E*3t^C)AQsHa@^h>|J2VpmY1D#8y#+5MjF$!5bLYXwmHys6UG?l4x
zx6}rkJH~9`L}mDd`MPxpnp(uzn1e-LB-6Nz>{$n;<(E#KJalWlNwJ0C2v=<-rAn;}
zGP1-$#;yTY`nB*jX!=m1AuDoQq;^u`!RvkS$l<4y{Cdm(6`j;D0sCco4BBh0+(g6G
z0_0V1sa_4CHx1zVm(-BFgLnnR4cf>~gOBXk!Xe}M!xyjvwiUwuE}1v;-G~+SX_e$`
zjA7fO-js(<y0I<*OW%Ejx5Dm-A;*k<J0;DDOJURg)gICsznJ+ur1qXCwVF6mZ!fNr
zsvwYcq2Y(GZG-6?I#@~jWcY4ypp;pUDI7y531XXBu_s(66OO=sEB-B276}{&?F*Xm
zPM7QVfcs9`*2&+~@QUt*BPqO`^#VR3L{c83V_W_B!Dg*l8Tr6xk<HS5Pbbzom~5#g
z;m2jpAMv0Eo6nHikc?O^-;^GT989p8IS3n7#yRE(Gj4R=l9u)IEge6tmF@!@(5;_H
zqfDD?<+lrJhtM?hb<gU{nr~uS{bY|T3i68$|B?2M$Fwfos8^ki1!~bDB0~ZXNbb8`
znRg(K6ov)w)jy9-F64UX_q(CoJ7t~R<#JZb4w^dFsgq~PucEC6W4&!1D;K{r@|+7V
zhPL@V!kgNE!TFe!?qj-MBY$pa8)H!emGZviYh^IUr>2CTNnGxr12?(+-DR#Q)S+PT
zQn&|qx}A6m@3eZW<GYWU^c+ddGyNp=Jh9?%r-3p|R{*E_fa|t}{<GD~=$}>{ydjn6
z_fhiFr1ogKR_tIS?T-znf{aU{R9aZlNi7~;iuTgqRz}~~3I8kH%Zc0>8TLXXN3B&l
zv)4Cpqk-GJ19=^nrZ_WH|2m&e()^|BPg%#dYOBWwXGku;b>20EZgSe3oAjTBQY{X&
zO+D7(y*HvW@8ecFHz^CcPUi!OWZK=17S^;qw-TI5-x3n1!h$=sD=v^B6|mt%w9YY+
zaWue*%aB&a>9}u#iS4hWY0~UN+=<~|+(A=;KdYn(RS0`>|2)iuww+SQiG@crjQcD<
zl2M=Ok2J#vv_Acij$b!7w`k+>cdJP}4xPT8yYW1gl&-TPIeesG*1kivZHdxLvB2~h
z>)e<1H|}UVHy*pKEunE-Rf|lI*h%-EBL-E&%N5K6)G!W18d9ZF`!jJ9&S~&neOJFl
zU^@jqYZ(T2elT&s)l(#8`4!gm4qieBUGA${)+=oePN%Aw;|by!vH>Ubn)IOlJn?#!
z4!RJBjSw)~X4;KH`TJSw4P*{-@}0GBsSI#w6E29AtOwe1mgP6#?OXtt!sTJhCf|d?
z9ciuWtiViN$fQ%KA-K0miC^1p90BZmBU+g&p+YyucbwZ$V=JIvjY+hxGO_I<(0Z&x
z30{|9FEL3|rwfUfw9Xa&uDMTt@K?Vx*;`wvcv_f7dY6+PwEUjZcN{FSSG6c%r%amI
zRukfBj$S%-%($otE&P51c^c<BnaQ|w)={R(0BH&X=7b~h8nm|Pm~+^&`zYs2B3&M?
zB|D^<=O?}E{h7D=i0EwsK`#>gk}d+b|B}HSc;GjL_JD2EH2gzk!<e-vE<i@@F_8Ri
zxvoPi){5cneGY()U&`6y02aY`=L04P4ZAimG*s(?u`Ua(n<R*>s63`uq{`Qgug&ZN
zlNYr3tjw5WApFN^xz2@40l9b>U{#qx^j<u{$-1H~cP_9Px0D7`fA6fo<IIv{P8>gS
z&eY9Ah+ZB3K*u|8Dpw)B@Hanck2D0DJ#*)h)MSvwzR3~Vx_OOJk8x=~8Hm>xi0Cz%
zp5`<6Qc-Cgj(+ja4Hg-)Qui+nlqN6K3rQ_K9IZv>z+)?NkSp~sf>~Ls$+)t^wjks8
zQfWN3L~3555Ym0xCV$^hc<2uuE?&9jjp1=L0sZlpSDNDysG)iB3Cp1){-2cWRS@qd
zA-E<vPolJSJ2x^<mu|WHCDH)x+M2*@IvZ}3tMITiMy|bR8iN|oa}N%up$Ldp0r&?m
zQz_brTq7KuUAh4dODmh@Yi$y`k05q3lar>|j3)J#46NaUY<j7hMHSDGvdnX6^Qd+C
zI->J881W!sr#*InqzbB4SZ?}tJ1jJTq{deyN%4Elci^hn<wyAx=$O^0J)b|Hbp`V^
z3SlW;s&@kQHN?1??rii6HXhT8lmPAJEWz5R?e7tDjv=^%X>B?4KW+tg6#dJv8GI^?
zQ;T9~APO`q!R+xPCI~&t+KNs7oIBJ(ogazDon6xLp2_$sYdIX<EIsLfn6YBCECd(x
z)Dhe}StT$1qUqCR)|r&*V@ptP-VJ|$ZA6eMOr3tWqh$J?iSkc|52)OD49tF~oi+rf
z0Qk^$DLQ{5(%4c>E`8U|ryJN)Cf&I|mtf4B;koSmi*r=j91wqc>ZjkFPn*b&_(<Jl
z&57abH&mu#LR6YbHYG{pHXRFq63A{2`_B85k=0xl$tLYMb6~Hj!u6>#_|F&mI)hi+
zx1I+-t(o=VE$M=f@Yl5UDG5b+fE?8`AhvOu8AfFU@sQ+skksG!$(ZHkNRH#9I>bVH
zT@O*4s_17bt<8*ED?{;SyR}5>ob?Y6OAHx`-fN9EInykUTK=aXORt~w7@mzCkij&~
zzh@?g63u^^sojta$(zuQcUq?3w{pj(^GCG&G|Iuv=To#_n_Y}aJ$J8TRxb(CO*C7z
z<CG_6;LMUe(ww2Kuc`6-q##vmHp0YqYngT)J<ice3D>+ibl|=;yrRx-@}UD7`^3Cb
z&*Owr`7U|hNFr5hp5&QXj{;Va%+C^NYUlUJ7?9o9X#eSqWr&ybC(;Gi>&Od&tfgBy
z#k`XM;3WTjpSd(2i>bW>sPFXc1K@o{V)cB|GFSP=!Ok8t;B}uirb){KH*NdE^_){S
zY4%HPeqcz_!`yG49!^~bu`LMu<`xHH<S`ezk7@R?NZ?$-y%W-nv#Hp4<|y6WP_n-<
zs*v27eoD6HbKOy`J0f}O;m)Mx&EOlcAHp#rR$hAo&-2qUUFOr}dn%6*jI{sJ)-)J2
zwK(S0YtHsf0LTOQF)6=G6`Lc`ep>!Zr2VlDKQd1H!=c1K7mN~VmHV{}>iBO&k+JUM
zEdtwatSim<5c!ER(r(??F;6gLaSm$Pe9XbY?wdB>berG52M7CY{{J~R*kcY3_TIGZ
z`>%t8{r?^u*sZg(J0VTW-%)S+jmiD=22;8qRS%6hRZq5g=*T8|Lus``jb>%?2zkm=
zUg^Fge_tR3yEAy~y^0|PXVo1fedc(2PQ@6PqOi@Bs4R>_5gr^x^~y}kR{AuQT&p%u
zKnX#y<etEZ3uX-Md@fmQQYhX`O>=ad<x=TKw_v)q<<XjzhCCtp&9_k6Hd3?+v(~uD
zY?2JEFvW$ZP$%^79`jHU&yV>a31bN`CF<#Ilp%-R$=q2T2)QN=k4$@9?yBEIq6o-@
z<)<@8khTUh%h4|kAz6RTli^=hh8Z{>+5f|;UwqF=Zj{M%YkQ=&2Ebg7&N=T`>n~GI
zY)3|x-a~o5;lxHm#=mhso|65+z|PVhcb$v)w-|T2Y$lk+#}UcX507tS$pHg;;(<=l
z74M_>St0!ov{LHB0U3o{{O|<BFyFw(5?X~^HaCy${P#=Q4gLDrfG%0{KFB=m{$`z*
zAD?B)lXxw5n+2Y9PhasmRh8;;_7-s|1hspT4kTkprHOU9A7drSGf5vHqZ@7>@#ES)
zI;>8Vav&XftFV?Heoi)(e+G4gv$W_$3Ih(P3%t*HEIDHk_hhWsxv<v;agF|72A}#}
z`1~_6I<A8r6zm!6qO@N;@gV~1Wnq(4aPE@763B1Jp|RB`NK!@-7ahDM#h#|CnUp`L
zacT(;=8h_vPCS^%!I^c=d)Iy!AeLH}w(bm*TD0e<m9hHnRV`l&I9ex4ZqM5oL+e`k
z>b!$VU<4f?r>nr(dSe6Ob@PYr<FB!^kMthX;uF%pA1QutT)Llo-(tlm8r&?n*ynAK
zW~U@vlab9p`Y5^2wf-9J1FJdbX-0eMIHDP`lFvSk-VK4M{Zkpi>GqJ7q?yAbe~IA_
z3Xzp`0x2EW-t!<#jiX7cdY-|T2cGhP*!h{EbI+kv$%}df&=@I0OUuupL1oMlO5^<F
z0OWvrMXqT$rV&<+TY>b@3&E(q`9{CKOCo;e%>=YQ4?2!qJmP8iQ>F7)Xbb1|W7Fq!
zO41PTtbo{FXA*hZv5jT5b-$FZl{#~p7oEwyWeKR`sC7^BpNDTkLqVJ;<(xEbzhct3
zh9m5TJE7*${=SPcuojISH;7w4zHxhr>uRl~en&kL0|B@p%1ws1QFywM;8!~D?iR%+
zI)3hgjz0WgSlmWMn$$|(7y<0iWf)VhS=SBN=~r0Ws#BZ6w;471=^#I@@ihk{zqBBO
zS^`!#t$A|k<}tQPhO&h+x(Tr4$(7g)-nyu?R^C0iXM4B-$!Ya5ZTEp%7%;q6I<Avf
zQ>JY6lELHacU`2W0Y2wUocy#iA!!46!1WzyRWv<w$Hj+>DM)mBrS6o>@%D=8ESt1#
ziOfF7*S8ZO*W?vcypM;u6Mnap*^3HW&Y03?$9<;NySw^1C3Hl7n4;Dyqb@kCc~Pux
z=wt+bFQ62TT1(zTm`ZJXg@LB)tBItPgEw+3{wlm@o-cnilTnxZC9_4<-Mw$qT*3P~
z3{B43DCTUML9_J97y0{uD=wj?O_v<u$AtCasCHb|ZknD3-jgZ8zNY)IsMop_cp1HE
zFlC4d0);Jd)XD9}UnL(&uUeVd$k=u8z~@)V(rI)f`-zS^yk2yR=7#Un`pdYIVpgQT
zPsqFrxjlgIMpEhQMgd!7osq`y(^|sQU`xg}$k(IS&3hYX@y2in-5BGob>cdQRuxnV
zj!P!5ZB5^bk5TEdW6M|WV&-LarmR<5kn|dbf<6;66eG2d!24)LqRZ~hhM^-K)^1Kr
zAm||!YfuYAP+Xq44cz{4DcU?^J9nIBf+w=nA_z4{#((GUs>SlVL4H4UNs<<GK-BRT
zQOPrn6%1;DF!>B7?F4efdbafvFK8@ILL<h<+1j3DvY>Y@F0Gf`tBdDhjy!^y+#AKX
zAF*>Yb!^%j4m@8Xz5CObgT;nlnj?0Y$B`>!;l-y%i<$J9iI29h7ov?@8>K75kMCzm
zko8FR7}G>K`%>M0Jf)iT5b$vmqhj}~vhX&9iu_VXcoT{E)&Y9EPA76wm}?GScs4Py
zVcEM^#*6(7|9QI}aLtf)W#c++awiHjY30w)Tyfh<fOsT=2_pzyIV4Qn&JcfW7Y6Cy
zFqoBtBNmK8muTmC8M?e(@{vF3zoo@)?A*v>&*{hX^E8>=R*IM8-nh2tHh-y%B7#5B
z`5x=lRU%!0RGQOg>B-bKA6S;hcG4@;2etSf5hqV3@%@OS<Zduw=y3Mhc8mtsx+>oh
zb{_dB8tJ7ajf4*0a2#V-CmbN-8WAhX%SR9C-ijr_Ze*`!6YpdAY+16~r8kT7M~CAL
z?L-D~>pK4C@b?7p;6|R*dh^$}0HYL)%f2*=)tm7dZED^cG@^@G6k0)n(i7(i8JjKV
zrIO9#q9_9G<G13VjxV0`R^THF`)@Akv0V0@zIm1Au-kHRezit^+`_UgpKk)XD~&Ka
z$T;UEG93%t1D7_ES67Yctx{}r(2~f1&-Z4vDk^j(Sho;~5%;4wlw$V-UeD~LQ9;r;
z=GX)|c&Xu>iE#>U{QBPB#JJ|}!4sEPKnNNyoMa_EG3yUFNcIJ73*>A0bwa9M>fCm2
z^m7k<d436D*byh+V+lO{7<J`~kGnYy9<d%Qr9-aq&@6LbVfx3zE_j9ED%byxc`>Pw
z0ED1bdCyL8@yWLj89mKunw9we_B%AD8IVDo)|Y9=A(9R|rg^|h3tkjs{GYS=X$UXa
zA;k>f3tH|4muF3$O>cNg%+W3sjrTuJqZs)l0%K*xiZyOAtPIei_;8}x0w=&)j=24>
zF&|eF`dvB_hm*d4#3=pi;r%%I1X!6qeq=^C$C}QYvqiVd|3JWY>&nBuD05b-;Xt$g
zUh0dSn1`u0ABuXRmaZ|bA7C1r{UEjL=`$R5zxks1#_j8`Cm}P;%7|4xX`DrCG#g@g
zY+ed&)shtCV#K<xz1}!s+B`qm$uO1Y?L{{6+<i<4XOlTP!_IvA18PMHRG75>L81rV
z*~3-JXtoyEyTm|oMC%t%sW^?7zXx$W%}8UcmPP`Nc~@%J=<frI(<GME%5=~>vU1Qm
zl?l$@iAkmix64$D4s9|+8clZ_#tdC}gp*%D|D5Z{L5-nSk<{KkX5(DF*NS)2zsYo4
z6b@cVoV177Wsqdhdaq+QoL*9x>7J-7+g_O$ik|zE#~jFkhj0yP^=;x)ZP1NMji+T7
zOsK;@A3VquYCe7IidxLDT^8f#MdfXhZ94x7$nfBIG^|;;YmNekZ}Uul^mZDEotOg)
zwN2*nPp@_2spcHv!OF1Z5P4^dhlN=QBn9t8T*3{&o!KN4XSFc^(}Mk_d-rh|$@_*=
zkDa5W@PzrY@h#=FE$Sp?=n87ZOUEhZoz#n*<uG1^5#;xfr9nGh+%i$_m*JzDeTs^L
z1G{)b;R(cO1xlFazUB##7a@LxbPeickQ-p1yg0K&$2se>QY6clpE!aPuC$!g4*3Js
ztR3LtHE${WV4#+tq|NA-v3p?6(2m<=TRHECF_#{8mx+wKKz*cNnIv|aHxB6{7jEgF
zurG$vGupjx9XVr;9Oyd=U(%m;F9TyT6zg`TFD)~G{{(DUhkF+ct8;Ep*?2fhJ~54T
zoz}jsb_Y$SWa((=2M^&PG`jx^3%}D76dKHfJjYX-`D_!^Dx|w#YCpX->X3a&yHDM5
z0Ca|C?ZX50tGOmZ>+Z$S)72T=%}K00caF}i-#Wa7v9M@cG>ujo;#GFV^&^-mVskF~
z{ZgoDD*LoI;KqpnFe9EL9L!Mm=tt0p5}ya3r6AhCo+h*Ypos&7q20QEPAd<3&@GMw
zyIRI0uwww&+9MpD9@bt50=<LS@~-t79Aw_2F=Ze=2wB3RIV(~%kJt17{wkAsAZN{O
zgSu!H`+U<Z^Yl^s12Y*2YJg-kk&~+~%0~B1qhez(Z^OT)Ra};K_rBKNi=n}MbC&#R
z@^QlO8@wMZl?R|4?d7;k^Cqhdz61AeX*rWgFn-g&XlL@i<d7o<c3(y^d0l8;0{%dj
zwR^_xMyutk5(^}xh!kzNUV3pxIJR>Gm~8>(=-|Lj+re~O;FGs;>9(LJ0g(~*4jY&J
mzyB`?|M_9@SC;v&6aVLbhD1g<n*T##XV33o{&fm6-~BHY%%xla

literal 0
HcmV?d00001

diff --git a/src/gor-predict.py b/src/gor-predict.py
new file mode 100644
index 0000000..d844f37
--- /dev/null
+++ b/src/gor-predict.py
@@ -0,0 +1,44 @@
+#!/usr/bin/env python
+
+from __future__ import print_function
+from utils import parse_pssm, parse_fasta, seq_to_profile
+
+import sys
+
+if __name__ == "__main__":
+
+	try:
+		import argparse
+		import numpy as np
+	except ImportError as e:
+		print('This program requires Python 2.7+ and numpy', file=sys.stderr)
+		raise ImportError(e)
+
+	parser = argparse.ArgumentParser(description='Use the GOR method to predict protein secondary structure from a PSSM or from a FASTA file.')
+	group = parser.add_mutually_exclusive_group(required=True)
+	group.add_argument('--pssm', type=str, help='File containing the PSSM profile.')
+	group.add_argument('--fasta', type=str, help='File containing the FASTA sequence.')
+	parser.add_argument('filename_model', type=str, help='File containing the GOR model.')
+	args = parser.parse_args()
+
+	model = np.load(args.filename_model)['model'].item()
+
+	if args.pssm:
+		profile = parse_pssm(args.pssm)
+	else:
+		profile = seq_to_profile(parse_fasta(args.fasta))
+	
+	dssp = ''
+
+	for i in range(0, len(profile)):
+		half_window_size = int((len(model['-'])-1)/2)
+		score_H, score_E, score_C = 0, 0, 0
+		for j in range(max(0,i-half_window_size), min(i+half_window_size+1,len(profile))):
+			for k in range(0, len(profile[i])):
+				score_H += profile[j][k] * model['H'][j-i+half_window_size][k]
+				score_E += profile[j][k] * model['E'][j-i+half_window_size][k]
+				score_C += profile[j][k] * model['-'][j-i+half_window_size][k]
+		scores = [score_H, score_E, score_C]
+		dssp += ['H', 'E', '-'][scores.index(max(scores))]
+
+	print(dssp)
diff --git a/src/gor-train.py b/src/gor-train.py
new file mode 100644
index 0000000..b8092df
--- /dev/null
+++ b/src/gor-train.py
@@ -0,0 +1,49 @@
+#!/usr/bin/env python
+
+from __future__ import print_function
+from utils import parse_pssm, parse_dssp
+
+import sys
+
+if __name__ == "__main__":
+	
+	try:
+		import argparse
+		import numpy as np
+	except ImportError as e:
+		print('This program requires Python 2.7+ and numpy', file=sys.stderr)
+		raise ImportError(e)
+
+	parser = argparse.ArgumentParser(description='Train the GOR method on PSSM and DSSP files.')
+	parser.add_argument('filename_id_list', type=str, help='File containing a list of ids.')
+	parser.add_argument('dir_pssm', type=str, help='Directory containing the PSSM files.')
+	parser.add_argument('dir_dssp', type=str, help='Directory containing the DSSP files.')
+	parser.add_argument('--filename_model', type=str, default='model', help='Name for the model file. Default is "model".')
+	parser.add_argument('--window_size', type=int, default=17, help='Size of the window to be considered when building the model. Default is 17.')
+	args = parser.parse_args()
+
+	with open(args.filename_id_list) as id_list:
+		
+		model = {}
+		model['H'] = np.zeros((args.window_size,20),dtype=float)
+		model['E'] = np.zeros((args.window_size,20),dtype=float)
+		model['-'] = np.zeros((args.window_size,20),dtype=float)
+	
+		for line in id_list:
+			line = line.rstrip()
+			profile = np.array(parse_pssm(args.dir_pssm+'/'+line+'.pssm'))
+			if np.sum(profile) != 0:
+				ss = parse_dssp(args.dir_dssp+'/'+line+'.dssp')
+				ss = ss.replace('C','-')
+				for i in range(0, len(profile)):
+					half_window_size = int((args.window_size-1)/2)
+					for j in range(max(0,i-half_window_size), min(i+half_window_size+1,len(profile))):
+						for k in range(0, len(profile[i])):
+							matrix = model[ss[i]]
+							matrix[j-i+half_window_size][k] += float(profile[j][k])
+							model[ss[i]] = matrix
+	
+		model['H'] = np.divide(model['H'], np.sum(model['H'], axis=1).reshape(17,1))
+		model['E'] = np.divide(model['E'], np.sum(model['E'], axis=1).reshape(17,1))
+		model['-'] = np.divide(model['-'], np.sum(model['-'], axis=1).reshape(17,1))
+		np.savez(args.filename_model+'.npz', model=model)
diff --git a/src/svm-predict.py b/src/svm-predict.py
new file mode 100644
index 0000000..6cbb8b1
--- /dev/null
+++ b/src/svm-predict.py
@@ -0,0 +1,51 @@
+#!/usr/bin/env python
+
+from __future__ import print_function
+from utils import parse_pssm, parse_fasta, seq_to_profile
+
+import sys
+
+if __name__ == "__main__":
+
+	try:
+		import argparse
+		import pickle
+		import numpy as np
+		from sklearn import svm
+	except ImportError as e:
+		print('This program requires Python 2.7+, numpy and scikit-learn', file=sys.stderr)
+		raise ImportError(e)
+
+	parser = argparse.ArgumentParser(description='Use the SVM method to predict protein secondary structure from a PSSM or from a FASTA file.')
+	group = parser.add_mutually_exclusive_group(required=True)
+	group.add_argument('--pssm', type=str, help='File containing the PSSM profile.')
+	group.add_argument('--fasta', type=str, help='File containing the FASTA sequence.')
+	parser.add_argument('--probs', action='store_true', help='Output class probabilities together with the prediction.')
+	parser.add_argument('filename_model', type=str, help='File containing the model.')
+	args = parser.parse_args()
+
+	model = pickle.load(open(args.filename_model, 'rb'))
+
+	if args.pssm:
+		profile = np.array(parse_pssm(args.pssm))
+	else:
+		profile = np.array(seq_to_profile(parse_fasta(args.fasta)))
+	
+	dssp = ''
+
+	for i in range(0, len(profile)):
+		half_window_size = int((len(model.support_vectors_[0])/20-1)/2)
+		part1 = np.zeros(20*max(0, half_window_size-i))
+		part2 = np.ndarray.flatten(profile[max(0,i-half_window_size):min(i+half_window_size+1,len(profile))])
+		part3 = np.zeros(20*max(0, half_window_size-(len(profile)-i-1)))
+		vec = np.concatenate((part1, part2, part3))
+		if args.probs:
+			probs = model.predict_proba(vec.reshape(1, -1))[0]
+			dssp += ' '.join([str(round(prob, 3)).rjust(5) for prob in probs])
+			dssp += ' ' + ['H', 'E', '-'][np.argmax(probs)] + '\n'
+		else:
+			dssp += ['H', 'E', '-'][model.predict(vec.reshape(1, -1))[0]]
+	
+	if args.probs:
+		dssp = dssp[:-1]
+	print(dssp)
diff --git a/src/svm-train.py b/src/svm-train.py
new file mode 100644
index 0000000..887a921
--- /dev/null
+++ b/src/svm-train.py
@@ -0,0 +1,51 @@
+#!/usr/bin/env python
+
+from __future__ import print_function
+from utils import parse_pssm, parse_dssp
+
+import sys
+
+if __name__ == "__main__":
+	
+	try:
+		import argparse
+		import pickle
+		import numpy as np
+		from sklearn import svm
+	except ImportError as e:
+		print('This program requires Python 2.7+, numpy and scikit-learn', file=sys.stderr)
+		raise ImportError(e)
+
+	parser = argparse.ArgumentParser(description='Train the SVM method on PSSM and DSSP files.')
+	parser.add_argument('filename_id_list', type=str, help='File containing a list of ids.')
+	parser.add_argument('dir_pssm', type=str, help='Directory containing the PSSM files.')
+	parser.add_argument('dir_dssp', type=str, help='Directory containing the DSSP files.')
+	parser.add_argument('--filename_model', type=str, default='model', help='Name for the model file. Default is "model".')
+	parser.add_argument('--window_size', type=int, default=17, help='Size of the window to be considered when building the model. Default is 17.')
+	args = parser.parse_args()
+
+	with open(args.filename_id_list) as id_list:
+		
+		X, y = [], []
+
+		ss_map = {'H': 0, 'E': 1, 'C': 2, '-': 2}
+
+		for line in id_list:
+			line = line.rstrip()
+			profile = np.array(parse_pssm(args.dir_pssm+'/'+line+'.pssm'))
+			if np.sum(profile) != 0:
+				ss = parse_dssp(args.dir_dssp+'/'+line+'.dssp')
+				for i in range(0, len(profile)):
+					half_window_size = int((args.window_size-1)/2)
+					part1 = np.zeros(20*max(0, half_window_size-i))
+					part2 = np.ndarray.flatten(profile[max(0,i-half_window_size):min(i+half_window_size+1,len(profile))])
+					part3 = np.zeros(20*max(0, half_window_size-(len(profile)-i-1)))
+					vec = np.concatenate((part1, part2, part3))
+					X.append(vec.tolist())
+					y.append(ss_map[ss[i]])
+
+		X, y = np.array(X), np.array(y)
+	
+		model = svm.SVC(kernel='rbf',C=1,gamma=0.125,probability=True)
+		model.fit(X, y)
+		pickle.dump(model, open(args.filename_model+'.sav', 'wb'))
diff --git a/src/utils.py b/src/utils.py
new file mode 100644
index 0000000..5235ca1
--- /dev/null
+++ b/src/utils.py
@@ -0,0 +1,39 @@
+#!/usr/bin/python
+
+def parse_pssm(pssm_filename):
+	profile = []
+	with open(pssm_filename) as pssm:
+		pssm_lines = pssm.readlines()
+		for line in pssm_lines[3:-6]:
+			profile_line = []
+			for n in line.rstrip().split()[22:-2]:
+				profile_line.append(float(n)/100)
+			profile.append(profile_line)
+	return profile
+
+def parse_dssp(dssp_filename):
+	ss = ''
+	with open(dssp_filename) as dssp:
+		dssp.readline()
+		ss = dssp.readline().rstrip()
+	return ss
+
+def parse_fasta(fasta_filename):
+	seq = ''
+	with open(fasta_filename) as fasta:
+		fasta.readline()
+		seq = fasta.readline().rstrip()
+	return seq
+
+def seq_to_profile(seq):
+	profile = []
+	aa_list = ['A', 'R', 'N', 'D', 'C', 'Q', 'E', 'G', 'H', 'I', 'L', 'K', 'M', 'F', 'P', 'S', 'T', 'W', 'Y', 'V']
+	for res in seq:
+		profile_line = []
+		for aa in aa_list:
+			if res == aa:
+				profile_line.append(1)
+			else:
+				profile_line.append(0)
+		profile.append(profile_line)
+	return profile