From cecbef02f6f9b88780d59830547a40a5cfd7f0dd Mon Sep 17 00:00:00 2001 From: liaozhaorun Date: Wed, 28 May 2025 14:16:04 +0800 Subject: [PATCH] Classify2-2025-5-28 --- .../factor/__pycache__/factor.cpython-311.pyc | Bin 163184 -> 177177 bytes main/factor/factor.py | 707 +++-- main/train/Classify2.ipynb | 177 +- main/train/Rank2.ipynb | 544 ++-- .../catboost_info/catboost_training.json | 1576 +++++------ .../catboost_info/learn/events.out.tfevents | Bin 43210 -> 43210 bytes main/train/catboost_info/learn_error.tsv | 972 +++---- .../catboost_info/test/events.out.tfevents | Bin 43210 -> 43210 bytes .../catboost_info/test1/events.out.tfevents | Bin 12560 -> 0 bytes main/train/catboost_info/test_error.tsv | 638 ++--- main/train/catboost_info/time_left.tsv | 1576 +++++------ ...ex.574dcb96-dc9ebe4e-fec52066-4d735471.tmp | Bin 1310 -> 0 bytes ...ex.6e271dc5-5022da26-a9883518-dbceac3f.tmp | Bin 746 -> 0 bytes ...ex.7083db21-2a602535-fc74a793-f24f9ca1.tmp | Bin 4 -> 0 bytes ...ex.c8182d04-ba2f7d40-17be2993-7ebf9fac.tmp | Bin 4 -> 0 bytes ...ex.d44701fd-1e864ae1-30a8fb3f-c689cc7f.tmp | Bin 4 -> 0 bytes ...ex.ea11afac-3656bfa6-7640c1ea-9df76981.tmp | Bin 4 -> 0 bytes main/train/predictions_test.tsv | 2304 ++++++++--------- main/utils/__pycache__/utils.cpython-311.pyc | Bin 5084 -> 5101 bytes 19 files changed, 4335 insertions(+), 4159 deletions(-) delete mode 100644 main/train/catboost_info/test1/events.out.tfevents delete mode 100644 main/train/catboost_info/tmp/cat_feature_index.574dcb96-dc9ebe4e-fec52066-4d735471.tmp delete mode 100644 main/train/catboost_info/tmp/cat_feature_index.6e271dc5-5022da26-a9883518-dbceac3f.tmp delete mode 100644 main/train/catboost_info/tmp/cat_feature_index.7083db21-2a602535-fc74a793-f24f9ca1.tmp delete mode 100644 main/train/catboost_info/tmp/cat_feature_index.c8182d04-ba2f7d40-17be2993-7ebf9fac.tmp delete mode 100644 main/train/catboost_info/tmp/cat_feature_index.d44701fd-1e864ae1-30a8fb3f-c689cc7f.tmp delete mode 100644 main/train/catboost_info/tmp/cat_feature_index.ea11afac-3656bfa6-7640c1ea-9df76981.tmp diff --git a/main/factor/__pycache__/factor.cpython-311.pyc b/main/factor/__pycache__/factor.cpython-311.pyc index c489e508b6eda4f87009a55943cde3eadec8871e..cdae86a6decff8298b93a7d0dc3be3344a95fb81 100644 GIT binary patch delta 31185 zcmd6Q34B!5x%iwrOJ*jMec$(x3E4Li_OOQ~K-dE^OfnNl$i|%{6a*%lXpxFvDXs90lQU zLq9j2B$0lumcqScDtVmy3CV|)-$BYr?jHc3;363;Q%2!>CWZV2Ti|}g1d=DY6O5J& zaPKf_P~sC>LM=_~r|N@*;NyTm!}r?8Hy40=t6Sv*Z+>+i^OrzcuHktMbB{o+ z7ij2J2)&3QRW;-nKphK>t%6t~#Lm)K4aABd_7ayYk0U>$Y3m`a1kzr{7}q9Ol2^Q_$xMM$tShYKL zAH-I=V?4xGxnq40TkVeB53w3|>;Z_a5n}-|!-Ei8YanYR+INQfm8KMOWhgvb#gG^- zF{eE?2$+oxM-YV|8bKTYdn)%{P9pbC&Z8uk+mTyG7IDAI)seiRsd=G95sbykx%v6o zq?T*V-$!IapXN^?Ymy-SY9N49#IEjWY%#VQ*~T^ly9tvN>}IUC1%O@Np>J-ivr8-2 z*4PzY`sPlfiLJ$49RL#nkj&~aUdO#LM@!V)y@i`dJ$JS6{gUmNyaNI4O@W~%1lS4N zjKFszq0mSHx2RYP4Ye0PHy5k3bS79+>oq*K1rqGimGjoKZCL0gBa7#bl+1+&|D)u@ z!Y(Y!B5;q#N2_aARNk`HRBp7i6j}`|>t9HRnRlyJYt~fVveis3VOp7gJv3#Pv!+H} zG57s}NN!2_-1Mc`d<+1fo7}Xcv7uuE92Rg)c`6Jnqr5)3+*OG^fFa*VFXD!2g=^)H z71LgpVJ}jJD@veK?ewSOusVC^#Qn|UqGw>) z6*Jx|TZzREA;4N%8EeMy4-sIO>|F?a4|^`RXXZR;=9QTz=F(xiDeBF}PCi*U35qV{ z-kepap}1G*j4gUyDM#k^WCY+4+-Ogy3`-0x-=!ibb}j%QHJPo|dtG7u$}!&d9L zYjf^To=m04YELDoY4@Y!EjuiEOb3 zR(lY@)hrA>gurK2|Ml2?KjznSk@Ih;9&wB6o4l$AFBo5)rA%>?-nV#FS1I2{>n@!{|!$T$VPz=QH8R_AJ$Btwt4FIh?UbH7?r z7)4nqKnK!dFm~zco7+srQ-pn*3#-ajh^mpBUv(lE+hUg@aE%?pc3FKJ%lbw;In0$U zEryXaEq!kat%zKPlg5ftQ)kNr^?JF3%TNQJUsj)h^|VA3F`vZcb}f&EvPYJGoJL2O z0u?Apv{Em@`~@~6g~{_P(qQs0uQ(2MR#o@MQO17->-cWHWMw9l+`MvM>hGYC3zu6b zu8_J4D^#s|y?_oU0ZY#G!<>p2 zu)miOyoO*HTlH@>?$*_}HFvhOn%KW#@>vA0B6uEvDFXmkxG{;V-sqR8hlo9NyRk#Z zwlz05wrb-$2-3OfU76ZM45k9GYnt1TAf^t`QC%#GfnB9H z?P+N-cCdTc8tzb6a!xxfNDJw~KF}MQSQ|EgGK_eEpD*XINIABI`)${?C-qu^BgyOIsRSSq7tvAgYzKft+SfOzsp+L7p?o0VKoET-v@p1Ubyq*wWr? zbn(ghp_)C%N!=uh=lALO!uLnYsTOh#65BR^a*-yc&n2&f%^hBGy4*(MyL45P?qEM><^ z4B^_bmR*3somi%s+kV%UsfaaL1P~pkq^uk5Y#}t2efHG`6*h$d_O?#kP`m z&emgs0f7;KU9IadwzP9^AB+x*gt$owjch{_zz-~lj)Re8?~q^5(+tTT8ali}uGoQ9 z8#zVq)@3NDH$le%Ea?lO!C23Xx!nkm=yb7uyLUm1q36M#0*CswHqO0impC8<6Qz|hy& zeHVr(TYE6;=Y?4uI>juz8$)}DBwWH$%=$s~ZOPl5h<*52EV*mwrDJNvToI#C+8(TH z1%_`1ZHmGVn!_jhJY48R!~TJVw?KpO{;3``tYydH!Osg1xWxFHZ%*84}Pr zFRXnK%OQ9UWW#hC0$+AZt_=k}@VS)i;476Rf{T7Q?lz>u^`PXeRn7W(keeedWt z(239~(HXx8Wm$><3x+5k(B-faXYFN6oW*P^94}$$X9!+KFlk;n^NT_cuZ%~^4NKov z>|crdBEB3b0U6pj-eGz>)U!WGfUVqu6_ z1KZZFqr3)!1nT|Op~uhXDns481$N_8*g+HbM`7`nTW}LC>B|VAglWQdDE`l3=nVvK zBA7ImFI#V7U<7zAthXT^zd_*EJ9eqQ&LpZ@ihdsmW_|KsxR;5bYgDy%7+KhHux}6b z56@S+mW-?RndA_c`JcrNo)zyr&W)}f?_!AJ;FlQs6#`#qI)oW?(RmL;|AFAw2r6MJ zPP=!~$V&c0$;2$_ZR45|PagK8xbs-MKJcVGrSaMrl15?$RkjQ$qE?*M?AGB$zD!>kbmK9{&r z%=tZnNn=mAd=pml7HEpSx3KMRaJUqE9{FB_853G3zWEJzm$;4xQU`IXv02yNW@7(y z=-w;$DO~(E^)E5mHP|1;7IlyNocW~Z`~F`FCgv7@FVUhaoJYR=aDr5o`Ct_G5y&-( zwz!w#35AV+Yg?C*gAZt*1|RDb+8c8Z&SHVnZ< zjCxjipNP1GWiBI_G)j(qQBd!;;>PQACrT)3$E|iA>ia^;4K(MHzKkXx50!tpQAT!f zN59I;qnoCCV64^by#QgWO09BgkfU~RDsc8HPUkB3=daeo7Jb3j%`*R0m?G+K{$)Z+ zE2s!j9`T8l&#*j|y1!xQa|Dyc4}X{;xs&4=54%uKp-7{0;{!q38@A4W!BKgvLXSZR z+h>25hJiJWuauGqa?P?qN^WP!bjxq#WCx=VQGz?6ADG3dVRLR2gDz7;f?d9&lQng) z|HRCPk*UxWxBFMMw70Px3t1N2Z8%ss(4w@c$8!HbKnF&v;lQZ%LRsCl$kKLo#_zSa zwW`4hM|Yd-SC$njQX@BemRYG?rQ6X6)-1|8+9UQ5YSghH{xB|M>-LTvBBd&^*SQGj zNyorhIZS@lz!~aecU`Y00 zv`Thw@9dMWvg``*gx2e-MazV;uSfbTka%tccw2dsX6%Y6qMdloPCLUR z`IM7AEbg{ASOsFl2aEf*GyJKYYiPo`v3{11qDZkn$>Af?NScxLoxNP~$}Iah$xHMTa`6|KhY@T{hX z>0obDO+b&tHY2{&9c{W+a4@rt^`e7{uKU6KLj@#Wu?736MVYR~Wd@;SOP*0SfYs}|l-O4gA{IOg~#%Sfrmi85{} zJEinm-TjfAg@{zQ5)?Uj5q9azn=%gz&$CoaBb!~;3ID(>qFv)_4wUtlVGYmfy&R(} z5O_}x$FTqt7Xqke7h!NRQ{6qAl-5nOYM;7c)Bf+zw+cs6WeJ=^Lj3~VWCKPz#OE#| zg`|boEh42$e9fLxklOe-Q_hFt z*@FS9seICzw=E{A>h&;62aMcwKfn41B+Zhxgs2q@-6*gJ>Wr{e!eba#4?c0xvrj4h zT@dy+)ugz5BEqsO5skjsa|32wzWIQlNex^o?w_ASP` z1;@I@VqZu4Vdr5KRTgai* z7WYyh3;1{Tl1z=bc;W08GI$YT{w=JmjklA+?GuUP1Pa|_j(WIlVug-_-Hp@TZE@UA zf)yTZ!_ut;C6KR;hS%h%Iemi^OoG3jx`Rxc*t~YGS9Cs}Ern-NV?~JClRg#yKAawc zFOZs22;jk3phT_~)Y?AFwjU6wqRCBHEnLad8T5_K^bP^L6^es59nQO`vkwn=>AvFo z*eFF#3Zf8;v3DT&F7T3fm{i=7?9}4)MeNDq#f-x>TYh~Q7Vi!2`Lk%2z>M)MrHwV{ zTkDNJJ(Ls+-^FBmkCHhP%jgmuq_5@tPRu}123{poC>kQ^mXc%SZuv6zTb^if^GLdvN?cmOu27FL0%?JwSd$=FpLUiu%FrmJmHZ z?4KZU%>^S4$=IZ=7&KbcFO2!-rIQ5<7%xUs#`D_oV@RSg)do{B4x|S}M-i-2^<+oPr zu|p)=FcGb^&Yc%(DZ3o{FKiz_J=}=jVz1eUbL_JW43QR1`VX)?+Js<6IZ&+esulYv z6!F-He8oAklhpH1oFi{(eOGvFOa*M>4FB#$aJ;uq%rIOU;J}&AYP9hgzF6kT|;9lDC0(bv+GXQyO#g>Ir$K-`dQAr zLyoBys$u+lnVe+SACW`tDYQjkIoqLw8(^I#U7L6}PpxZnCNzt#LdxXSWcdw$r$BNm zevK`>kANP(2pF;=sDuWcuEb7^Pk}f~9jjW5X7is2q(nUx!#Mz|`9BNbjapzO>uzD` zyl5pkjpKMG4_eR$d6~~s1n+JQ^G)B;*T01rDhsp#mx;V5J0wAd7J*B;nb7;wK~t zOsRm)%-LqpKJXNwu(i=U3Skpj=*;uZ_)N;E)X=0#?Vvy@Q6|{@QXLYxJn92wfCHLd zzzDT1SZ4vP?}DD1akT0+g(NHcWLJ-$5S(UJrjG_i_5?aK{O>;@>GDxcV!uCs?rjnw zcc=s8QGC>=q+~hVOqIz+09^@zxdSNx;Mc0mvnlfgW!|MJ+Ho!R1iJox2|o@MG=R@g zp1eb1q!(hQ@Ja8GXrkcD-yw6St0e)ULg)@*?ka@z*Q#l>X&MDh^TF7bjdJkLSA`0N`i?l!(gSt}-+@ z)#6xb`JDgjHBxSVK?T^)gwn7Q3c*$4Z(rD;ba0Pt+7h7ve|y8kCqVI9t-mN3RBWiQ zcnoXGe8}sfZkNNwL4Akl$6W-yxb(~=EO8kKC}FQ)@M8pz^F#k2wF!}+WLO}ssA;_P z4Bo=!v%er2pkuT{C(rrazjH~XmLP)k&yy}&CeBg(8{W!9Mpt1{oF%pc6KDyNOA z!+KZP)TtwC?XX&VCc~;Ou&E0^lM+?o1<0zlsnbT(*~9AWGo@B_p-o-*nVeA-nL*Ix z5!Zqx*=3{35Fs?xrc4tuN`cw4QS!1C-Bm(hhE~)eF{uM-DMpABs zWT2!%q72~wbe+_INiyeOOi$Kb=f7WjcL168Aon&0 zY;U;x#}e=#y^Z~CqWxV^s_#`FR`*s6MD(bg6biB_=YFUPJ{;8>awzr)_ozN8+E4IH z?pqB2eyy?;heo0a?5<+#`M|HqpUiZ}=Nuw-s_N3SjkrJ5!5d}n(b}buEZzf?@4&(!m|7o!X$6V$595? z)isJ_cB&LO8K!HUdyb;RUWyV-6$2U_VAs;~U=(S)hQ5v@a*PRVi>(Y(7^KIJRWNM| z?3RZari(E3&LL`LjDjIae1U=~PgsdX(I5kgFURma1kUkVeyCtRA_17~LcEzbDVe2k z*7cf_S;qvZ`3M#BI5k=EhBrT~Vv5sHacUWM8@80_rc5_AYuGwS-J8gNq+}ww_L)j1 zPR%b-Gs(c6jcO)-sb}UMOG_n+0G)IYqJlOG4FlVQ3WF!SOwriV2nXHbODd_5_l4zs zH3QqmdhVO17>l2VIU^$#{0qU%P(m_fgYSkH;4&M=AwUE+T3L*aW049PbVN6;sWdo`W5LIx) zZ2NXm2Iy&_oLcNCXGbpPi1wcRhev4 zCilnKlv63^I~XQ20v37z3U^2)Q3re`@l%5Z0pN>Zi8|tN+@XXc362QVqVW>7=16pJ zst}nuF!StsLAAiDT3}Nx=$;1)?5uERcH||QN>Eo?W%F#Zd4g=-sDA)Tyye+U<{!jz zD3W=Y38aEg@qkkwZ=&u3I*na)&+vQLwQRa^Qx00?}=vu3*+hruhpJ%eMj z{*U7TXNv9f-gO&V)>CE~9ihPQ9D*E}DMfDxsCKO;4?=-*? zNaeqrPsw1WD3BKR6#o5eCdHqf4~2dpl4EZM(_{uqL1gUZ{kcL|fmK;(Qx*#OYFz5o3gC1NEDi#%%VegD z+`hpo+h~(*6l5E{xcy6+$mbLxl`!_n@OexQQ{Zp)CVvjIKi0$Q?#4t`kLD9|m}mw# zT%OCUnM;w3X!6D3>@>(Q$pE9c{SnpRA}ovGCB*9*zAu-_jC3%v2nE>jPSCyLnp&a> z;@`_E)7wG?W7L*4BX{_PypTY zt^q>9cfQstTW6E46J+aLX#B4{<_m?_QMi)*4gbedW+vIgCzUZS? zj46t*hl#_D;L>pRxp=tyx z5g^;Mdt9>0uEI3WeIe@pn#@ImlyauTT<8y#-TsRk-cAV>4F!)Tqb>?F4xj~qU9q;N zvU>4CwiQ$Ui79^Im!R*PQZFFAKokrT$9{<=sMCY4l+LX;=tS3bxg|J_Dg^GEa&{G~ z->oyjrL#t_w&VhRrmysbs=uhe#gn98zV51I%9sWI{7);HBKo=~ z*Z@>T1oMBcWaedvj2w(~rHt%hL0ZbghqpfbF5+*mWVHNOCZaNLx;aNWPuIqAr04P) zM-Gqy;-;HH5-aMa458Vp&%VBgTbnqd2g(G{#bN37? z_>uWcTDV*uL|5+m^ZtUSUi{_f?jxy7WTEm^L0?`|a-Ti^6U)|x%py1CpQ-{wwt|1R ziYZSXuZuh+{CB8Vz*jC~;`y+DkZ|*zlcL_I$TZY4^5_!}3Gc#O5A~dP^|vAQC<)#A z!AmK?kpU_Ng{Tzxnah~eP$SgXhV+t+p%;H-8B5Gr{)AY{+t|GE)m=$#GBzN}@K|9b$i z58H-e&(6DBT$1Y`b#Db$nG%*av$+l2!=~8~(6`hX^t0dPPi|zgL+i0B3=H=c(zebH z_G|vaMy9->3HpP58+W77CTk3OQp6HiWOfJQ&D?fYva`%&$ z9(?%HtH&da(b}nO^?cmq`>a>|kQ}yBe9q81d6j+ovR7>fy2S0)2;m5er9w z{4fo@5?;8^Y~7yCe@}%?I9n;XZG(J$h5{6+TshX%9tK`&n|OsO7H4EG8z=# zqY>2`mHBdLSzI$Ba1uCE`Y{T7fET5T3RoNgMTR|kqQTQk*suzCgw5O82-@HC<|2ze zH|qazX%-H8AER0B!{H!UjGBdk@@QX4qGOLA=&u`yw`G=Dqo&!SrghKj-e;ACkIEDW ztM91pt?Jt+gyjqxx~r|S88+DrK{mr7lglF34M)ln8p|S zGX)hcr>k>^?<&v+XGqk6+ei9$^+Z`!TANDyp(^ljM9&_pD#oUY;g9$;(Nj{avQ)ZD zKe+gg#l1QqJY%4vd$CnkVw05!vJ#gT#v5~CqE0E9Sf|8Js8c-Jgv&pj!auZ`3FmS? zpAn@3*;kL{YS1M9E=_`k6$XvL@3ayX&jR1Y13;yL!Z7%wau$CeM!mFRY%gr6ASIBN z;|52gTl@_HRQUys?c1q$f;+Q-dR}_x$|GkkpLirWs}wYhHu=)ABUgX?I!!1lpedJs z^w{M?ccZd-=-{R9S3seFPJ)0>-IFxt#9#sql;fp=#+29T>+6lpa3c;KcDho~MowL1 z9mLXL}-PnvxV?E=fhEAgQ{(^YC%>#Mr8c}o08853IACRiS=m*{M#~{ z>RSl?QMe2mkLe(13<0MD>GID#G88P`FWERns%EJx;}IY_91-Fd8|v^4U-yv~iz>h#E-r^-r{lWB<#BIOUq7KA5A+gJ zy2_fxhEYejEQ7JFd8`8WD6sI|DA(HnHQkcYgw+1h6I%ur4(8c%Du)v)Va4BRRmP5H z*(vg^1!!fhY)u57!nPrXHV2jx>60>DEW{Wjut1V{h zNKDgkOq0;Gixu|l7h;;MF$Zih2f&~>V3kEWWKwydL*fQ%aM9lzdpNeYY9OO0)~YJ7 zsR~9^Wy7kn!Tna%LYrz~_dKU6>D>zV7zPZeMv85+VnJ5yLyKhIM2k#rHz8#V3jSFQ zlZgsJwO?XTg1TTt+dxWOd2_!B=C@uae$}eZ?6JDw(yJU=|E|S0xy$6-5QrK8<()3R zmj|A@-1F4chaS6p&#_CrKl=8oaK1|e)Q?~P;FVxlf#HjHuJ?jH3vu9Bl|a9G2p@ZX zc1;-H+`xc%d+oa1;++rTEvNPgY|+NpF>+6AFKn@)oWi9Btv%3tMTb|#@L~n~I5PN2WP6B7oI8Oa&&@k(ckG=_@nLp;K!9F+@P!bM8W`8W zqbNK^Q4yqN4J6spDn`;Ohtn$0hYFi&g|td*+GbnY=8?2*!)e>BX?3==I+XV$IF}tw zoiY$)OD#WJDil`M*k-J;rmn%fM5{8%sSAX(5+St|wLzIJs7z3nxwU~vi*`3H61-^v z!=XQJN^l?D5p5=bW5o?P_N!)30g~);c(c@GTm;S~1$?G(2YhDwDQLrDhI zKO!F0?G1?cXU+Q7o#KrM)86TmYSn{u*zvjQ4j{tkLKmQUD`pFGR@y37jZ~}~u2?7B zW)Rvtg^G37iY{A47oxJ%sw{J%(wtT-q?RBqOKm}=g0ht2626V*%gpn>m+%xa(lf|N z@L%HERI)$CWCX`?L{HEW-AsR)Z?;R4%BP{ebAp@bJF``_*rr+x2a+;IUIqu8E}(Sj z$mIj;P)8QqRK+8zieXj7*{N35BAd#85hzLVU|zf}B3e}cuXW*SRcC}xT__qvk3EFC zP_&UmT`2B9U?bwzg~_$KCU~vHnCw#>M(bPJI$Jw#WeEWXiUoDxOKon17dr}FmeBu2 zwI^=hZmIUb$w4PE-rxy9&F36;VS(M>8N~0N+(hYlT{p&(dvjcKH+OhV-Xm=u6^F%N zLH5>(0+*86|Gkq*BPr#>Ddp!FVby9OrQDiQV@s(SN!c)*vca0N$(FJS44^hRBNMi79!qRNC4#a9EX;2;C$z&Lu|5=Upc$X4wsjPhQQw zfFg_y+QVHa@y3X#)uiHfkDh~|Y6Exa5LQvO81=36lQVWDdW;ZPV>E=_e=lWI%rQ6Y z)%xnfgLF^XhMyzGyPywHap5t2xwi1Ek?`E%@LX$nzAZezM>QH2-Irhsn=%qMbvSIQ zH7wT_mfNEo^$+Xab$C|KtkIm@o`phamQ|w_G};TfMT4`Ne;up1tRC_sgE>OLv1wH`a3-U6BJ`S=XnF z(DDG4b?nHvG+jf(1e$D$%c6r#5PbFs?v{46!QX)xsYMJU6b0PJ{KQtWFGr{I00eKr(<8>!)vskuV#9Jr2sZr(YwZNUbicB^&4R%>>M z4!8W7R@sy>OOWV?J3O%;ZgSQ*`u2%}D>$9X@Bf+v@rSFKh-iZ!c;u90y<(s#RBn&l zfgufF{Vsgb^bdE3^M5^0GPrZ^MOOH(B|6;T56O7x8F0gSVzBFd<2;E~7=jF%(*fu! zC@O^Bf}Mu2%2N_kJ+Zyu`G%AbT2AyTY?l~9AH$=Bu(5-8y7a<3=a1t-(XT)c4KBWH zx!nKC#j}Smzc?WJ4S*A!jt5%cvdiyq_31wJJBY^+u+pL{f;t*hu+(0Bsqe~@&y3^O zbl_KdCDqBNupyp+HS@e!Q@rpDW${be_~ju3ycN2`2m!eB>*e(Gx#)IttEW$u!5M>$ ztE`9vKMWl5@a=M@$c(~RExOE?B6}i6sZ#({79L)DZntO&JtOeqU3-8ysLpnHy=sqX z?+jr7TMpBssj;ZpsB_l-1lhlYvj1psB=3LB|7gJRfPOL@oH7!WJsgxhm~t*e2+Fnw z&9?>39|>AA9JIt5wA2=~R8TGzH7z|r_IUpT-J_#h6wwnhZVO~qQ+EVd1FrJ{+~C{O zCo>bk(3>c{KSzMoZgzM;9SPLcImVx2b{|W=0S%0CEWCv+BKQC?vkG|i-|>BVyFMK< z0)IfrA@PC=NqpiX(;t}5DGmjh=M2ZqITv7!tG2~ekHpmu$JGk@I-#*eh^w{6wc6rZ zQRhTiWznN>S7Yz7z5QBiRE{kwXJDNzs(2)7#&Fb(b49|MwL;VkYZRod8;PnNj;ggr zZLvjd8Hw649JNDeYZrFkE=29HM(wjj?QC5vnlh?V@o9DJeJa2fBfK6*q2|%}#CW z))lUU#b4mmP_a1=V%hUJ5Cnfm-S8IGnm5}s-)zh5K2nsJF&HSxFMOuysiw0r*1Y+) zy!q!=+Va+pidqDXc znZHp8**az$o}FxmJqJ7>2g^3ovd#2#BikyQijpaVyYAjw>N{^DjsNx`Mbb?UDZp@$ zb2HV^VFrztxhfta;^E+S@q8lCryUX0nKD*nMW5tNEZQ*#7~YP`aUz`RWtZ04k0BT4&*llB8E!UaO)$Q8_LIMmw@T$x-$5A|w1hkDwnLe6ZPb`Gp>i)wB2 zHw#tmJ=PgG0xl-r zDL#6{#}Q}VJ>s)heBhT6peTMp0H!H}DTuD}haiD{7{Mb5`Z1rdzr=^kS24imMJx!r z0P!XjBp^b>J0FXI+no3|i*K845K|~hDbQg85&!7ZI=dXc5eQ!oz-v2pMZ3P$pf`#7 zx&_H!XT^4D};#cBW zpew8~NLKoNFcwKFe$EQXNk6hAet!rF$s!pk5m_W5{k)0z;Sr=A{iX$_8-4GawPPB- zBg=k;fB-t_TZQ5~cKEg#OJ5%{mtl~;XMyid(A(`S-a%&ZVk?W+L)k6_c!5y7nn!Q6 z;Vl|^6M&@$Zq#W^-6nMZOl@VlC{pcm5vf# zX)k7HXO0~*NnFNgc=BjevLj#%ArTG9kp~ae!v*=>X$L0-Zf!NIUg(yh9413vxdgvm8R; zHyQ)NA=<~B(9raOF!~q_eLjACIa)1TNdkL|j?Z*RAkaUbJ_hEUUJe01Wj_9?BWjj<02?=#b5IU!Y!3)7*4mr(% zj|fCL{Af_5Aq36N3Su=Ugty9^jpjI18Q3NOXOj@3En>3RAcn;DFoZ?Dz*-nz^k`(% zXw(#k3cic#kN|Kt1)&RO937@BBq8ArKMaX&VF=55f%OFbw3b6M7Az&Xm^c<}Bc+%) z7PJ#JP_O6&5Cj5)MbBd(o&=to_ZH9ulE#BrzGOd>XI?;3fPSGTfyxjVOnp9w2F8OW zB$xya6rF}vA#iryE6^we#)I?-7(~s9Ee;6;2I@{X)4+I;7vsu|F+*eQEBhQ0NE#2y z5?zf!Kx_;G4#`+Bm-xd3>P|ve5Ex{ihsGc<9>n8-YEFwIJ6rdPb7YM;P_OajVX@TH kIW*~P(JM2(@;pTgf}qQqlh7Ch2J4=O#vm{upbG8(0A8Jk!2kdN delta 21637 zcmb_k3tZGyzMpgE!3=Lv1crw^#X&(qMKnb~Q31vGTN!4U0Y-trGl0*8c5AcOYW{2g z*4oPU#MX+U8t+y&+pO%~+Wte+mFBLy?RMSlwNb5`m0S1sJO5#3#39)}OVjs%{^$LB zoZtDK$KQJ{gbseAi+Us?B2)o?1^>)#JaL}f85Kv{w0BOb2oo%t8_A|nox*b0o%^!L z-FM!fL4F~uP{n4)+^F6(#v)pDR^_%ZEQ^1ZP;0m)+!|4@v5bZE7|RjN8xjlQv6g#y zI1a+&ETS-nOdz7LiQ^tJncOQpOr}E2|3J(yg?|7Z6~s$uZ$ongLlp;#oqKX{4KHDOKuqf@Ox>^}{4iR|u_U0u}dTvaSf;vf%9j-lIA2 zHqr7RPniqhY)E;Crxe56B+J7X79vzy@(7PDgV-EP56{04!jmEYad=B6kMi^-5SMHD z4TgndDy{Hq<#^!*DE$~OQ4Oinpv2?+Z3VpL!`l=5Z56x~z+117s!kO0RpZs`AZCWJ zRFxpu)mqYrxk5HUT%qM@3=7BAG2|KHd36tUlv=QxBMKQOW=7Bo)touHdolIEmD)!3d0M1mvxRb($LpB zSw~5{uqU@QHWK8^#v+VIh(pi|-{+ntQ-srbW>PL>Pc@R@Px0#x3W{Y$C@C*qg-CLjALv=4aA9T%RA?wXTHcO%cOdxs6QtJF%Sx|V zYnm`))=a2%+pO+#-p$-=)w*hR#Wicq5_T6ALan_;wV4%IVLZTb;!!&qZS`#b&bfa&Z>17jQF=mK>)MQUji#0cE87B59;Ktr zGD01*gcaA#fI4ov?$Jan&VD*_dvhUgqHu6-0^~nF_l?48%z6L;OBz&cKZb8az-HM^ z2*JBPUwCI;8B`Nre6)mj-6zLD<@IVu`6%SLP)M1d4;8MMe~=>{%C3mtAqi>4r41IhSGaR)pXdhfI`J(M>MD$Y@+ao zvefaMgTYz0N|?LgI-qdZf}=Td2`qLmz=xACbQB?I(Xh4H;4w@$3vA(4#bd8gd_zF- zrt;y%8CQcJXnk`)@%lxVBDh&WYha*ki?DrhuBH}VJh6gnajej{csgknK3F^}1VzK1 zRERH~Cu4=yIcj0vlCtP0u^92FEVlZ3DE^dia7hBG6^<{dARWS&OEOWVDsCN@0@;p3 z?N|kBAloBHA-OEm(>}-u(;)SL75ZF zyA!#tKZj+4*M4ZlM9Aq}u`m4{$TY;%ts@uMzY+^PwDQGiygNyllM|uz(1tcyCa*&z zh>S>(*rgYxl0R zXi=jK>FgungLUcp`{2c+a@ZPL*s~Z6X*TV$HG}YP6f)LN3X_!q;@IoLy7db+oV7Oz z-Rmd8y#M<8gQQB>xZy6cNch`^OtM%QyK(;Pr?I`02rnSGu-32^>rP{xy{Wyq#lb$u zXeYw+2z>yKECAuojfq0{MomTu{IOby@x(M(+l_s-r6`J znuPX!X4_7Qxo>mR99>urlwivdRteqhv7Ku$ScKI?GpobiVrr?g4poBE_vo7J4f2SR z=l>~Kpbk?a>P&6M4ttZ)(Q4anWqh2yjHzhbV5(FBlN=t}-pV}?81eRk;}ovPr1$za55T z^szm$K)S8NYBaa+#%{GYL(ta1#t3OUlS#NRe`h+05X?I>qaTOdT!R=lyBWeSc1|=T zV`LiE6W(M;IXK#kU`!Zlzei_s>~3zhwz1u8rEq>{>J&TA$TOMRn9*d}#@@yXQUTE4 zg1<}iStdLfOiH^h#(jptK>!B@NcDPbUkLMdZw*IXTf;73jt{TY%Vb$EG39Rn4kbWp zY8RI~`$G8V?x~$$Vj|a89s3Bwe@D28fNIOa&~|?ZA&;`z*1`zBmcy$-%_1P|QJb1t z8%>^Y3$wTSv~MEjLKSqV0aD@5t7~IW*BrLy)+Vo(C>dMZ_xPT}YEZD!u2hnQWE`g+ zjm6esvsgVLwia+%ZI)^7KCC93YdlSGz4y!Y}iQ^zM{f0u9zhlLJqHC3oy71^K2JhzIn@JWF<@r z$fM&*=935K!f?h`9@x-{JOyLPdHOMgztJtSwb&euR!izPB+n4ebXfO4AVs!_$YX|P zkx@+GP88Z~kFI=Iowc>iW^a)t@;&C^BGB^2nHlrfAy@!Bp~g0AbE}YcOMFf&ygRgz zzZPWxfAWZrZi0c1TH0)H2ARR^wY#tGmfuif=!?32nOf6;bu|j+J8KqkPL0f;>oI3A z4o6`p_hV`&fLIbk6NR+H3*(F!zXzgZJ@%@iN!WLINi=edbzzw%gl6G}-i7I0gYEd< zilwg9U?{L>l5oS}R3Z0uZRZIP9NU3ec})yMIK8ko07GO@hTnv3a)z5QCpW#}7`luJ z?HJ<2z|ZuM0Hz}$oawBCC{h$`7rt|*hd}Y#)LK8!$3Zv8QFc>b>|LRn5}Cu0+Y@bU zfgv#@5U@Kz+38nPFKdkVmdk^CB-RKo-8;F{$BZ>>7&k%!xPfz0Fj9}pL)=GTAbmKl z{nW9aA>jy5feaicAaJQmaj|dvFTYfh2Ell28gU4Rj>Q?c&ifb&V;XIUtiY(dUh#sO z3evGea!~l>SPL;fu(?|~q?4uxax%G2a?{Hd6D_Q4DswP2Y6gvYFn8<74ARI9;#v_z z4cM*N-CmyEu-%8b&pQpIC7W0 zGUe>Ogq25Oe|YD8>KnA*cVoW0`(Ar=n|ei{X7c%o;VRPDVrR|X=?-g7Mer+SP}}LH z3I_2tH1#jN0h&5oQN4DO%YoA~%wqZna1bdS4;-whU^q?{Y$w7MV>M;SLdpg*f>T~+ zZ>hsI0Xx?B=#z(hX0gPXq32@r)6a!EYI^_Z%=v+lsS8j%PV0W;^@w9sGi(US6#9wN zKxsZuy|zERmD%fUI&@CzOuN~`eTPoI>m$wnpA)7QVRIu}{8UU1)Y@yyH|50udPk7& zh*JGwZGH$O>9~qSbSh2FtwvbKIoJuvQr`F4i@B6k^j&=ED<836yk=a_3E?Dhil0Hn z;0$;cLtMaoCM!XzanySK0w*SOtv`vO|3LUH!l;!%_j*Eq@p<>Ti1{#AeTU$)AReX3?2wHn=XVf}va1wM`19L(a=P#G+Y7WqX1@FnbIAeW zCx4h;@1LYzLn^q%KZ7Ao?dur&FN9zWIe-a#qWEtNy@Bv1LJ4%*tA?Xy%f53nM%K1K zS9Cve#IISLOMa&`2uBQF6U)`>hfHkLl_qw)aKpLrl)crr?vF`IRUlJFE;=MD_Ku-CBuFbH?!v?VwB;HR+QFr&&fER$&3diqBL2v86#;xqBT%b=a>U z)ggPzsrUzO-;0Csp{%VN8sle`T}9JToeDQB)QjQ+c1v#S7Lk&1!0@vszvXo*ek!~@w+~-@J$;sI$rN5 zsDKCXks(?Nts z-Pq2cmP?p;A8H>u*`Bb9=2kmvD`yPW)!2VHmg3?uW4?oxql^5TRCV3;vJkXa{VEx0S z|0BjrGVl4UK1b6 zB!whO{6i+`B;&;`StLDbHm3Xx%d1Sx%tTifiH)2I=U=#GWLyNZXrkDgMT!#3Fe5+P z>cQOJ3`-!x1W`GWq>tz3H!dKp4ARn_*G(i3k!vW)KeI_KKiq*`3DGi%q;=wW;!~nW zW8KweZLxTiuu&K=sZaiea!sM(gAkU~~_=W)Ej&-nEln)0r zG_r=hh)p$#OJ|X_gzvpwLyXEIGP7YMr{K6Yz$pkt@lHdg`s}2v=;hdCu$`1r%#3gx z`sPqkv1C2TaQ?0cl-uhUi2o=ihSkAH;+k2B75E*4Dtujv5J)R_YXL?s1gK`^7#!uU z<7Dw~b4Z-{*?f{<9+3xEE#?1pMl8WTEpcuuAu%BlK6!$M9T0C>O!7&S_}pSLb4f5h zaMAKr1kx;TZLY-XRsnd$?V~<`_-ly$o+V_Cf8T8Q5-U^|`$yQKU)bEvha$CF5(5Iduh@ zUNn+qSvB%8nBuO(gmun)SCE_4SM=4pfNnN1um$qeoQh2Pcl7nOb!1lQzuwo)*jI!? z*g^WLirra2cmxr z`bQ{^#^c`YcNPMJ)%=i&}ZN}ar{0jMcPS$SV>kFtL-?ftKm8<+Tgvv+M+@0nf=Ol1qa)2V3 z*m3gxw+z<2c2e{s49gBAXNU7w?PP)`$=`96vBA{p(X}$G_mI6we0MLIKK^Bt19$cV z49o&CeIJ<^F3)pF&q6AGeu6Bxh5(-4NAgFU=ncUwzTe;VYum^cA3NKD1a>%=UQeO} z_BI$_t8kG7Tge0v$NhEReD<%+=DFfaog_ibypbe~?8$ox<&^nXFMjf_98BCt4LU{a zMM4q2Ld8@<;1ZmbMJm*xQmEmGSiPRaI$bxCzp1zR#0=}bP@~0UYvMN(*j6Y5N8q-W zHY>wR3ve>b*D+mKGv`$*aul<%8xTgtrfX=yt+$cc*VF>;x(1W|XqmGgySv|+b34rS z>wWYybYp;GJdv?83tln_x@XF{@b4h@lfz_w*@!dhV(etF+5e}QfKUj#T8YK zzCBGb?H;0^%=hLpQAYoc&c1dJnf)Uy%FWo>o1I1XlCN->#@`}qRw_$mN4j`6&f@RFk%lDCObhBS@NFT#XLZhFn*fy+qJAnT<7;Cu$qwjFO z)J-Z0X>x|&PfW^rzLs#6?{@?FZs4eRd+q5xUmSBkNfvV-CJ8@+%LlR72c7#LCVxRL zn|esRD!_iLxo0AeQht}JNxb_6Da`VZ;s7dq)@}6O`pU=1?2%2HckgnTw=SeQCoNO`~RLW@OUkzHfx&=wiobJ2QPv%e!I&P3Q=&dZ*th{ z`CR4Ez_y_YBGFMaMDVVQ+fI`V@$hM~7xq3PUm<#O#%Xwk{0XiDI(28rVO^FE`n!it zNoAoB@ZaHs>8jBP_r=;BMmuW|Pq|2J7*}cTlJtx3xk&EpH?hd?5%|&0TNsiMN}#9S zy~s9vp9t@a@61lZSA)2ypUlwYV0g0F)lb^5p%^i5lPSXoK`xXwL_!|OHNoEd}8BT!Z$Pcu<_{&u*|p5>j^V4ybNbVhI+!x@FKqOHql1l^#hL@7-MW@ zkz)Vbq_RNXV1jr@1hiY58t^CY2S?VKn(Eq{K#WG*vt#c%tKHsfk=?f%h`fC{GYnfP?$Lo!pP!qAxyNq$LG0u1WJa(b#pe#$6q zX$*Xo!=Eb@(TgbDr38eS&_#4Wl_-Am3Yn#^CFi0NBz=ukz6DWVG&asdjtd zD}>w0VsBbt0#Kx@L5ct&9`L=8BzH*qphBsRKc|a?xah@H+S)1y0uhRaF2SP7@PGus zbCYsTXb!|lV=`U3tbypb11W=%V%{etQ!kaRk=B`{QnQpa&y`f?PO1X}1~n1tcyY%k zWJYKFphBgV0a}%!@_O!7O7~Oe~8phlsMmlE^(L-M4Myn*nf z?nrlPp%jk4fs`p{Q>OQ)Oqb>@la`yMxph*?bXSVSonkpqbD-v;MiDty8f%kES0eJC zD`J~FVw)7P?Y-!Xo?`dpQYjjL=V#4+y5mralsertX2zvKts*qJyV{*OM-q>JN^;fE ziEzF1-A~D{wJ_ZF{)0Ra^F9tJoEczqP@aIqX5Kv>1OqO8E)M$2x3S^8w3;GatnMLCQxCj!JoQQNyKDQ=7a8T@I}b znZ*CY9OtPbLc#uu!4DAb7sp&8n>4Y|WA>T2;}RJUbIgHDBrE10kh~Oox)1`;)4@sN zicH3$F4b$a3De4@HOm*YQCRur?7@JPb z1CW0%-DbCLv!vU6E^pe2`U45l*c_K`@<4Qgq)WJ{70*GLPAs$eR55`3bLqCYbz3Ce zmLO#YLlyC9-6`(4X=meR^~cS6LG6kwcE=T;igm>;I+Gx+u60+K&Q>(`S2Vt3YjIWB z-4%9MoLypjLH(~MRX7+NWGVd$AFJsX@aTjm7`-)gBZcweAe-booA51u6O~lsE@~Ay zg}pYZWQDtMrL?-vWw5vnmaYmZGRvi!2r3bq)!p1%BaL0ywG>DbBM(RVgbab+eE_cP zT6cEsp-5M>$sKL#sz61Gg}Rwp9Q=m7S@AH624`x|ShL-}9kgoKN8FOyZ^$Q-;(J_}z7u|sEW;;sKA ziJjczZujkttxm>NvL6EVq_9@sY~=D|>dITH9Ue z%B*r{R-IYy&fIb~)6$=5k*xKu%m#O6!`aNH{>&yUyDMQy`GFm z*I{Bd{&@s^C+8#6ql9SrhQ#;7=?nFhXkjWK?J*<`c>|2pCmqYVH>Z2!$(cuTT>4VC zzVwvAtzUXpzouWmMq0bhHAlbRtzUmuzqwz(*`?p=)^9zlxAp67QtJ+B*FH&abLp>l z>#y%B>*_p|-__aG=~BhLCsTso5F|W9>Gw1VHDM8hX;H!WEfPOHY=@Yvq^TrNELPI^ zVm@8_sFDvGC=P9IF-&Ndbh3Z+8Pwqk=igY#s^tJr=q@GAkE+FHv_L3_r<`sj?I55k z*d3=uO$h~mw?2dxCG!u1(3Lb$R)wWX5xkvqJ{Lm&Od@!)m(%;j<2qW&O`#`A%nhYU zVr(d#LC@>Ns!;j>HTyp@6SZM2A$YZ6iBK)e#0+Nf^jKPzI1$5~ zPd?Rr3BnJ)f!Pt_!ErPh zCV|Js(Uj0>P}_H?@L!4VjH7cp!Qkq}J%{&nPjSUfb;nKZ)w$#5oQ*5)k1HO?m@8$> zl@U&i;^%*^P{wQijsRP1Au&JMenaz(&4VEdb?n7(MQHf#@dMh>0d4$1SlmE(%%Dyk zs`(st-=G9QSJ_~+LLDN7q`6h;XH|xNm7yoCca3;^ES(HK7T={Sb*akSsxnDcHZ=A{ z*S&PCSbLPHoqvs`dsX3l^m86%E0N)g1DxzuklJu3+oh-3q)hx;PvLVxF*lL6#QF>f z^VS1Ao+|bx(pZ@CUrnU*(l~dKQH;CUd|v(m|(=DZ9#JIscEUg6^gmw2X(SKlNP17LOotp?=_tyCrf{; ze!9`w{8aOk_NVP#KStHD+OQqiy}5GtY*}<^;v@_k5Q-6GGcyH)^F=9(uA9c?z(v4~ zKrYXP!<{d3^6(WecMMsALUG+hkan%OA0DBlkh~UUQXoDtk#6hMD$+6@3B5nGXX(jY zY4Q?RYK1$s;*7zay5?-^hW^wI15?VSDdjSPwx?|*pzdj??rFo-Jq31^ed?Yz5FURd zR|=aHROe&3&gZyQ;?ZncIHc(7U8)Uk)doqmA-JNSBhEh>C{@bp&bc#xRO-GWmre`2 zH$V?=%ca)`XuP*7VBAzOIS;I%U7VXoSC?>3BgcYiJevs#4i#uSF7sqty%6&v{0v!K zD!!FRC+Y^q*(ClGk0jB|fKjl5p5dcm1!{ZL5aK9U$w$FTJ_=UyQBX6L-ffWd^Q8G^ zDWUE_oJ(hM>nvSm=Ttf=bd5{3)~#AAsn&WK=ggi)ztRMp;A>gGIC&PGOLmEkvuI7a ze>ES(Jb=pW$rJ6f=|nNIh-&F?L&f+aT99N0TH*Tj%Q(RKVO0Z$aMET`Vto;vS3enw zR?9IsA<8=~{;ZYA9vdc9V2D>#g`uSg%MeiA*)Bl&jG-Fi{1>M9X5DCRi)@=sXLSAv zCGOi+zv`o;7};C!8#BJwvL8Fp1mMxESyfuSxSXMT$1Y(^2yCVBFKPLnCBH!xg#!=# z>cD%q-YtYu;FkCL;36brF@o>rphw3{JB8 zy9*afYigueqbt_vjx`>j2MTWq9n^>iuA{%K)+(|MJ<0BgvrZPAtbcyan<1|2mb)gd zkXEgBOzrRqV}?F$64#f}P1msfzb~VO^wUsLvw#-F zaLuK%O5vT2_Sxe0BB~cH3us1`thD?p2Ul9Zy78H2)H)Zjkk0D7{KGouJ=Ghgb9@Ez z9V!v`wYX0C_MwJ!Xk8E;3PT+lx-5bFPUGb70f(v4*@ZMSYzvhCE2>c<6TMpV=X;+9K5$U8{_)s#aN;A3BOlQ;T`_t-OX^rl*M%0XCmp0|S z@$p9z?rQF4Jy{R#?iK&HoF*sqG@lvc+w(Z9U(v5$;nJ^i>sL8Ho=ek-bM9jLwlc@R zp|3MCD=x;2%RCWy4;2n#nrv;mjmx-z+o#I1U0p>JQimHPzidY>)=#VG%+3j6Ae@m+ zN$^u?Zc6;Jy$J>E7pwP6h8h$rm#xn<1lR-E_eXo6@!Eq&meTaFIw(P*KMu7l$gh^t zQn+zBdl}6u(<&yUbtfEcmiL8vE8koot$@Y*R>@fFT3zc}YH}|%@l7JLJH>o9rM^F< z-j&kmP67EGnD5di4r-h|%cza0aLpqwKT5_sCoiYJRBF)5U#t=b$Iukq(g4a=_gR>RaxXcaLJ-1tj?pVuY zVVcqbH?3^gt_(HLMLXR66yWX)e(q(REa)Z5rn^ZJ}YB=;2h0 z`^@x#v=*!rX=Qc@UAZoV*gZ4F8_YCAtgE9j`g>4ETd@FE&L$%Oo%o0qR%d4CD^@y(Earw7X#tJ$ z%ZqR2<0omjoI_#l#x7yDA%AQib`;?W>`J}3sDZ9Xd1vhg*PM;+IU8LQHe$<(E>)6u zF9M3nO5fi=zgMPn{d*r9kT*F!#K!hx8}L80fiv`%+v!yLT9}yGM04gM-#zMzmbStf zY%k>D=j~7yokSj%WBIG!&F3=uR=l-|rlfI_xF63CEBwy9_gepM@%1KJ5nYQF@)i(& z8|F-Jrj&GI`vxNGx{}SpOd_@}pUP?2sQF@?)?3z#ANp6K_EnItja3e!We8+#x>>$`3();CU*L>6gh- zh+y|2{0iY1rW5u$K2*W70XN^rjB4Z&Rs#{p5Z1}LRt^#wVmRN$S}RBJL=Iaqg2?}7 z&QgyWu2{nfDw1 z5K!9e?+D)_5Ug2)Fax2KqWn>~3@_6%6fD2*$*)avk;-@aP@D{fDPO`uK{6C07b2G+ zzvm&}z(CpYV}5=P&d<%*yBLSt_Uu~(1+bIvSZhc0e|$Aomk2PV>kvtsCZK^t?vxYBXD%+gIbb_?>?x8f436eePAXA zeD}cyIP-}aQ~-3BJqRs9;3SX&4{7}-`PmF7l0%sSLXB{nA!e{s0octR+&74UQ$Qg+ zq)l7pXRCuhheJ^swh18f5&|b7BRr%v>*eRRX8s%wOGrBQJ^&WO`x26iu~$MZnU0ZH p!ZxA>c@!K47DJ%7;AuH df["cost_85pct"]) & (df["volume_ratio"] > 2), 1, 0 ) @@ -2161,8 +2158,8 @@ def calculate_arbr(df: pd.DataFrame, N: int = 26): WARNING: 此函数会原地修改输入的 DataFrame 'df'。 """ - ar_col_name = 'AR' - br_col_name = 'BR' + ar_col_name = "AR" + br_col_name = "BR" print(f"开始计算因子: {ar_col_name}, {br_col_name} (原地修改)...") _temp_cols = [] # 记录中间列 @@ -2226,7 +2223,7 @@ def calculate_arbr(df: pd.DataFrame, N: int = 26): _safe_divide(sum_h_minus_pc_pos, sum_pc_minus_l_pos, default_val=np.nan) * 100 ) # BR 通常乘以 100 - df[f'{ar_col_name}_{br_col_name}'] = df[ar_col_name] - df[br_col_name] + df[f"{ar_col_name}_{br_col_name}"] = df[ar_col_name] - df[br_col_name] print(f"因子 {ar_col_name}, {br_col_name} 计算成功。") @@ -2251,14 +2248,15 @@ def calculate_arbr(df: pd.DataFrame, N: int = 26): df.drop(columns=cols_to_drop, inplace=True) print(f"因子 {ar_col_name}, {br_col_name} 计算流程结束。") + def add_financial_factor( main_df: pd.DataFrame, financial_df: pd.DataFrame, - factor_value_col: str, # 财务指标值所在的列 - ts_code_col: str = 'ts_code', - trade_date_col: str = 'trade_date', - ann_date_col: str = 'ann_date', # 公告日期 - f_ann_date_col: str = 'f_ann_date', # 实际公告日期 (优先使用) + factor_value_col: str, # 财务指标值所在的列 + ts_code_col: str = "ts_code", + trade_date_col: str = "trade_date", + ann_date_col: str = "ann_date", # 公告日期 + f_ann_date_col: str = "f_ann_date", # 实际公告日期 (优先使用) ) -> pd.DataFrame: """ 将财务指标数据（如每股未分配利润）作为因子添加到主时间序列 DataFrame 中。 @@ -2295,47 +2293,60 @@ def add_financial_factor( effective_date_col = ann_date_col print(f"使用 '{ann_date_col}' 作为财务数据生效日期。") else: - raise ValueError(f"财务指标 DataFrame 必须包含列 '{f_ann_date_col}' 或 '{ann_date_col}' 作为数据生效日期") + raise ValueError( + f"财务指标 DataFrame 必须包含列 '{f_ann_date_col}' 或 '{ann_date_col}' 作为数据生效日期" + ) required_financial_cols.append(effective_date_col) if not all(col in financial_df.columns for col in required_financial_cols): - raise ValueError(f"财务指标 DataFrame 必须包含列: {required_financial_cols}") + raise ValueError(f"财务指标 DataFrame 必须包含列: {required_financial_cols}") # --- 数据准备和清理 --- # 确保日期列是 datetime 类型 # 使用 .copy() 避免 SettingWithCopyWarning main_df = main_df.copy() financial_df = financial_df.copy() - main_df[trade_date_col] = pd.to_datetime(main_df[trade_date_col], errors='coerce') - financial_df[effective_date_col] = pd.to_datetime(financial_df[effective_date_col], errors='coerce') + main_df[trade_date_col] = pd.to_datetime(main_df[trade_date_col], errors="coerce") + financial_df[effective_date_col] = pd.to_datetime( + financial_df[effective_date_col], errors="coerce" + ) # 确保股票代码是字符串类型 main_df[ts_code_col] = main_df[ts_code_col].astype(str) financial_df[ts_code_col] = financial_df[ts_code_col].astype(str) # 选取 financial_df 中需要合并的列 - financial_data_subset = financial_df[[ts_code_col, effective_date_col, factor_value_col]].copy() + financial_data_subset = financial_df[ + [ts_code_col, effective_date_col, factor_value_col] + ].copy() # *** 新增：处理右表合并键中的空值 *** initial_rows_financial = len(financial_data_subset) - financial_data_subset = financial_data_subset.dropna(subset=[ts_code_col, effective_date_col]) + financial_data_subset = financial_data_subset.dropna( + subset=[ts_code_col, effective_date_col] + ) rows_dropped = initial_rows_financial - len(financial_data_subset) if rows_dropped > 0: - print(f"警告: 从 financial_data_subset 中移除了 {rows_dropped} 行，因为其 '{ts_code_col}' 或 '{effective_date_col}' 列存在空值。") + print( + f"警告: 从 financial_data_subset 中移除了 {rows_dropped} 行，因为其 '{ts_code_col}' 或 '{effective_date_col}' 列存在空值。" + ) if financial_data_subset.empty: - print(f"警告: 清理空值后 financial_data_subset 为空，无法添加因子 '{new_factor_col_name}'。将填充 NaN。") - main_df[new_factor_col_name] = np.nan - return main_df - + print( + f"警告: 清理空值后 financial_data_subset 为空，无法添加因子 '{new_factor_col_name}'。将填充 NaN。" + ) + main_df[new_factor_col_name] = np.nan + return main_df # *** 修改：修正排序顺序以满足 merge_asof 要求 *** # 先按 ts_code 排序，再按日期排序 # main_df = main_df.sort_values(by=[ts_code_col, trade_date_col]) # financial_data_subset = financial_data_subset.sort_values(by=[ts_code_col, effective_date_col]) main_df = main_df.sort_values(by=[trade_date_col, ts_code_col]) - financial_data_subset = financial_data_subset.sort_values(by=[effective_date_col, ts_code_col]) - + financial_data_subset = financial_data_subset.sort_values( + by=[effective_date_col, ts_code_col] + ) + # --- 使用 merge_asof 计算因子 --- try: df_with_factor = pd.merge_asof( @@ -2344,43 +2355,52 @@ def add_financial_factor( left_on=trade_date_col, right_on=effective_date_col, by=ts_code_col, - direction='backward' + direction="backward", ) except Exception as e: - print(f"merge_asof 执行失败: {e}") - # 根据需要决定如何处理错误，这里填充 NaN - main_df[new_factor_col_name] = np.nan - return main_df - + print(f"merge_asof 执行失败: {e}") + # 根据需要决定如何处理错误，这里填充 NaN + main_df[new_factor_col_name] = np.nan + return main_df # --- 清理与重命名 --- # 移除右表的日期列（如果它与左表日期列名称不同） - if effective_date_col in df_with_factor.columns and effective_date_col != trade_date_col: - df_with_factor = df_with_factor.drop(columns=[effective_date_col]) + if ( + effective_date_col in df_with_factor.columns + and effective_date_col != trade_date_col + ): + df_with_factor = df_with_factor.drop(columns=[effective_date_col]) # 重命名新加入的因子列 if factor_value_col != new_factor_col_name: - if factor_value_col in df_with_factor.columns: - df_with_factor = df_with_factor.rename(columns={factor_value_col: new_factor_col_name}) - else: - # 这种情况理论上不应发生，因为 merge_asof 应该会把右表的非 key 列带过来 - print(f"警告: 合并后未找到原始因子列 '{factor_value_col}'，无法重命名。") - # 如果 factor_value_col 已是目标名称，则无需重命名 - if new_factor_col_name not in df_with_factor.columns: - # 如果目标名称也不存在，则可能合并失败或列名有问题 - df_with_factor[new_factor_col_name] = np.nan - + if factor_value_col in df_with_factor.columns: + df_with_factor = df_with_factor.rename( + columns={factor_value_col: new_factor_col_name} + ) + else: + # 这种情况理论上不应发生，因为 merge_asof 应该会把右表的非 key 列带过来 + print(f"警告: 合并后未找到原始因子列 '{factor_value_col}'，无法重命名。") + # 如果 factor_value_col 已是目标名称，则无需重命名 + if new_factor_col_name not in df_with_factor.columns: + # 如果目标名称也不存在，则可能合并失败或列名有问题 + df_with_factor[new_factor_col_name] = np.nan # 如果 factor_value_col 就是目标名称，确保该列存在 elif new_factor_col_name not in df_with_factor.columns: print(f"警告: 合并后未找到目标因子列 '{new_factor_col_name}'。填充 NaN。") df_with_factor[new_factor_col_name] = np.nan - return df_with_factor -def calculate_cashflow_to_ev_factor(df: pd.DataFrame, cashflow_df: pd.DataFrame, balancesheet_df: pd.DataFrame, market_cap_col: str = 'total_mv', date_col: str = 'trade_date', ts_code_col: str = 'ts_code') -> pd.DataFrame: +def calculate_cashflow_to_ev_factor( + df: pd.DataFrame, + cashflow_df: pd.DataFrame, + balancesheet_df: pd.DataFrame, + market_cap_col: str = "total_mv", + date_col: str = "trade_date", + ts_code_col: str = "ts_code", +) -> pd.DataFrame: """ 计算经营活动产生的现金流量净额TTM / 企业价值因子。 企业价值 = 司市值 + 负债合计 - 货币资金。 @@ -2400,18 +2420,18 @@ def calculate_cashflow_to_ev_factor(df: pd.DataFrame, cashflow_df: pd.DataFrame, Returns: pd.DataFrame: 添加了 'cashflow_to_ev_factor' 列的 DataFrame。 """ - df_factor = df.copy() # 创建副本以避免修改原始 DataFrame + df_factor = df.copy() # 创建副本以避免修改原始 DataFrame # 0. 确保必要的市场市值列存在 if market_cap_col not in df_factor.columns: print(f"错误: DataFrame 中缺少市值列 '{market_cap_col}'。无法计算因子。") # 添加一个空的因子列并返回 - df_factor['cashflow_to_ev_factor'] = np.nan + df_factor["cashflow_to_ev_factor"] = np.nan return df_factor # 1. 获取经营活动产生的现金流量净额 (TTM - **注意这里的潜在不准确性**) # 如果 add_financial_factor 只获取单季度，这里的 n_cashflow_act 将不是 TTM - df_factor = add_financial_factor(df_factor, cashflow_df, 'n_cashflow_act') + df_factor = add_financial_factor(df_factor, cashflow_df, "n_cashflow_act") # 如果 add_financial_factor 能够正确处理 TTM，那么上面的调用是正确的。 # 否则，您需要在 add_financial_factor 内部实现 TTM 逻辑，或者在调用 add_financial_factor # 获取多个季度数据后，在这里手动进行 TTM 求和。 @@ -2419,86 +2439,103 @@ def calculate_cashflow_to_ev_factor(df: pd.DataFrame, cashflow_df: pd.DataFrame, # 并命名为 ttm_n_cashflow_act 以示期望 # 重新命名获取的现金流列以便后续计算 - cashflow_col_name = 'n_cashflow_act' # 获取的列名 - ttm_cashflow_col = 'ttm_n_cashflow_act' # 因子计算中使用的列名 + cashflow_col_name = "n_cashflow_act" # 获取的列名 + ttm_cashflow_col = "ttm_n_cashflow_act" # 因子计算中使用的列名 if cashflow_col_name in df_factor.columns: - df_factor = df_factor.rename(columns={cashflow_col_name: ttm_cashflow_col}) + df_factor = df_factor.rename(columns={cashflow_col_name: ttm_cashflow_col}) else: # 如果 add_financial_factor 没成功添加列 print(f"错误: add_financial_factor 未能成功添加 '{cashflow_col_name}' 列。") - df_factor['cashflow_to_ev_factor'] = np.nan + df_factor["cashflow_to_ev_factor"] = np.nan return df_factor - # 2. 获取负债合计 - df_factor = add_financial_factor(df_factor, balancesheet_df, 'total_liab') - liab_col_name = 'total_liab' + df_factor = add_financial_factor(df_factor, balancesheet_df, "total_liab") + liab_col_name = "total_liab" if liab_col_name not in df_factor.columns: - print(f"错误: add_financial_factor 未能成功添加 '{liab_col_name}' 列。") - df_factor['cashflow_to_ev_factor'] = np.nan - return df_factor - + print(f"错误: add_financial_factor 未能成功添加 '{liab_col_name}' 列。") + df_factor["cashflow_to_ev_factor"] = np.nan + return df_factor # 3. 获取货币资金 - df_factor = add_financial_factor(df_factor, balancesheet_df, 'money_cap') - money_col_name = 'money_cap' + df_factor = add_financial_factor(df_factor, balancesheet_df, "money_cap") + money_col_name = "money_cap" if money_col_name not in df_factor.columns: - print(f"错误: add_financial_factor 未能成功添加 '{money_col_name}' 列。") - df_factor['cashflow_to_ev_factor'] = np.nan - return df_factor - + print(f"错误: add_financial_factor 未能成功添加 '{money_col_name}' 列。") + df_factor["cashflow_to_ev_factor"] = np.nan + return df_factor # 4. 计算企业价值 (Enterprise Value) # 确保参与计算的列是数值类型，并处理 NaN (NaN + X = NaN, NaN - X = NaN) - enterprise_value = df_factor[market_cap_col].astype(float) * 10000 + df_factor[liab_col_name].astype(float) - df_factor[money_col_name].astype(float) + enterprise_value = ( + df_factor[market_cap_col].astype(float) * 10000 + + df_factor[liab_col_name].astype(float) + - df_factor[money_col_name].astype(float) + ) # 5. 计算最终因子：经营活动产生的现金流量净额TTM / 企业价值 # 使用之前定义的安全除法 - df_factor['cashflow_to_ev_factor'] = _safe_divide(df_factor[ttm_cashflow_col], enterprise_value) + df_factor["cashflow_to_ev_factor"] = _safe_divide( + df_factor[ttm_cashflow_col], enterprise_value + ) # 6. 删除临时添加的财务数据列 cols_to_drop = [ttm_cashflow_col, liab_col_name, money_col_name] - df_factor = df_factor.drop(columns=[col for col in cols_to_drop if col in df_factor.columns]) - + df_factor = df_factor.drop( + columns=[col for col in cols_to_drop if col in df_factor.columns] + ) return df_factor - -def caculate_book_to_price_ratio(df: pd.DataFrame, fina_indicator_df: pd.DataFrame) -> pd.DataFrame: - if 'bps' not in df.columns: - df = add_financial_factor(df, fina_indicator_df, factor_value_col='bps') - df['book_to_price_ratio'] = df['bps'] / df['close'] - df = df.drop(columns=['bps']) +def caculate_book_to_price_ratio( + df: pd.DataFrame, fina_indicator_df: pd.DataFrame +) -> pd.DataFrame: + if "bps" not in df.columns: + df = add_financial_factor(df, fina_indicator_df, factor_value_col="bps") + df["book_to_price_ratio"] = df["bps"] / df["close"] + df = df.drop(columns=["bps"]) return df def turnover_rate_n(df: pd.DataFrame, n: int) -> pd.DataFrame: - df[f'turnover_rate_mean_{n}'] = df.groupby('ts_code', group_keys=False)['turnover_rate'].rolling(n).mean().reset_index(level=0, drop=True) + df[f"turnover_rate_mean_{n}"] = ( + df.groupby("ts_code", group_keys=False)["turnover_rate"] + .rolling(n) + .mean() + .reset_index(level=0, drop=True) + ) return df + def variance_n(df: pd.DataFrame, n: int) -> pd.DataFrame: - df[f'variance_{n}'] = df.groupby('ts_code', group_keys=False)['pct_chg'].rolling(n).var().reset_index(level=0, drop=True) + df[f"variance_{n}"] = ( + df.groupby("ts_code", group_keys=False)["pct_chg"] + .rolling(n) + .var() + .reset_index(level=0, drop=True) + ) return df + def bbi_ratio_factor(df: pd.DataFrame) -> pd.DataFrame: df_factor = df - + # 确保数据按股票代码和日期排序，这对滚动计算非常重要 - df_factor = df_factor.sort_values(by=['ts_code', 'trade_date']) + df_factor = df_factor.sort_values(by=["ts_code", "trade_date"]) # 获取收盘价列 - close_prices = df_factor['close'] + close_prices = df_factor["close"] # 1. 根据 ts_code 分组计算各周期的简单移动平均线 (SMA) - grouped = df_factor.groupby('ts_code', group_keys=False) + grouped = df_factor.groupby("ts_code", group_keys=False) # 计算不同周期的 SMA，并使用 reset_index 展平索引 - sma3 = grouped['close'].rolling(3).mean().reset_index(level=0, drop=True) - sma6 = grouped['close'].rolling(6).mean().reset_index(level=0, drop=True) - sma12 = grouped['close'].rolling(12).mean().reset_index(level=0, drop=True) - sma24 = grouped['close'].rolling(24).mean().reset_index(level=0, drop=True) + sma3 = grouped["close"].rolling(3).mean().reset_index(level=0, drop=True) + sma6 = grouped["close"].rolling(6).mean().reset_index(level=0, drop=True) + sma12 = grouped["close"].rolling(12).mean().reset_index(level=0, drop=True) + sma24 = grouped["close"].rolling(24).mean().reset_index(level=0, drop=True) # 2. 计算 BBI = (SMA3 + SMA6 + SMA12 + SMA24) / 4 print("计算 BBI...") @@ -2506,13 +2543,13 @@ def bbi_ratio_factor(df: pd.DataFrame) -> pd.DataFrame: bbi = (sma3 + sma6 + sma12 + sma24) / 4 # 3. 计算最终因子 = BBI / 收盘价 (使用安全除法) - df_factor['bbi_ratio_factor'] = _safe_divide(bbi, close_prices) + df_factor["bbi_ratio_factor"] = _safe_divide(bbi, close_prices) return df_factor def limit_factor(df: pd.DataFrame) -> pd.DataFrame: - grouped = df.groupby('ts_code', group_keys=False) + grouped = df.groupby("ts_code", group_keys=False) df["cat_up_limit"] = ( df["close"] == df["up_limit"] ) # 是否涨停（1表示涨停，0表示未涨停） @@ -2558,6 +2595,7 @@ import numpy as np # 假设 df 已经加载并包含 'ts_code', 'trade_date', 'pct_chg' 列 # 并且已经按照 'ts_code' 和 'trade_date' 进行了排序 + def daily_momentum_benchmark(df): """ 计算日级别动量基准 (Positive and Negative)，使用现有的 'pct_chg' 列。 @@ -2578,34 +2616,37 @@ def daily_momentum_benchmark(df): print("--- 计算日级别动量基准 (使用 pct_chg) ---") # 确保 pct_chg 列存在 - if 'pct_chg' not in df.columns: + if "pct_chg" not in df.columns: print("错误: DataFrame中没有'pct_chg'列，无法计算日级别动量基准。") return df # 计算每日的全市场动量基准 # 对于每一个交易日，计算所有股票中 pct_chg > 0 和 < 0 的平均值 # 使用 trade_date 进行分组 - daily_benchmarks = df.groupby('trade_date')['pct_chg'].agg( - daily_positive_benchmark = lambda x: x[x > 0].mean(), # 日级别上涨股票的平均涨跌幅 - daily_negative_benchmark = lambda x: x[x < 0].mean() # 日级别下跌股票的平均涨跌幅 - ).reset_index() - - # 将日级别动量基准合并回原始日线数据DataFrame - df = pd.merge( - df, - daily_benchmarks, - on='trade_date', - how='left' + daily_benchmarks = ( + df.groupby("trade_date")["pct_chg"] + .agg( + daily_positive_benchmark=lambda x: x[ + x > 0 + ].mean(), # 日级别上涨股票的平均涨跌幅 + daily_negative_benchmark=lambda x: x[ + x < 0 + ].mean(), # 日级别下跌股票的平均涨跌幅 + ) + .reset_index() ) - # 对可能出现的NaN基准进行填充，这里用0填充表示没有对应的同向基准 - df['daily_positive_benchmark'].fillna(0, inplace=True) - df['daily_negative_benchmark'].fillna(0, inplace=True) + # 将日级别动量基准合并回原始日线数据DataFrame + df = pd.merge(df, daily_benchmarks, on="trade_date", how="left") + # 对可能出现的NaN基准进行填充，这里用0填充表示没有对应的同向基准 + df["daily_positive_benchmark"].fillna(0, inplace=True) + df["daily_negative_benchmark"].fillna(0, inplace=True) print("日级别动量基准计算完成 (使用 pct_chg)。") return df + def daily_deviation(df): """ 计算日级别偏离度，使用现有的 'pct_chg' 列和计算出的日级别动量基准。 @@ -2625,22 +2666,24 @@ def daily_deviation(df): # 确保所需的列存在 df = daily_momentum_benchmark(df) - required_cols = ['pct_chg', 'daily_positive_benchmark', 'daily_negative_benchmark'] + required_cols = ["pct_chg", "daily_positive_benchmark", "daily_negative_benchmark"] if not all(col in df.columns for col in required_cols): - print(f"错误: 计算日级别偏离度需要以下列: {required_cols}。请先运行 daily_momentum_benchmark(df)。") + print( + f"错误: 计算日级别偏离度需要以下列: {required_cols}。请先运行 daily_momentum_benchmark(df)。" + ) return df conditions = [ - (df['pct_chg'] > 0) & (df['daily_positive_benchmark'] > 0), - (df['pct_chg'] < 0) & (df['daily_negative_benchmark'] < 0) + (df["pct_chg"] > 0) & (df["daily_positive_benchmark"] > 0), + (df["pct_chg"] < 0) & (df["daily_negative_benchmark"] < 0), ] choices = [ - df['pct_chg'] - df['daily_positive_benchmark'], - df['pct_chg'] - df['daily_negative_benchmark'] + df["pct_chg"] - df["daily_positive_benchmark"], + df["pct_chg"] - df["daily_negative_benchmark"], ] - df['daily_deviation'] = np.select(conditions, choices, default=0) + df["daily_deviation"] = np.select(conditions, choices, default=0) - df = df.drop(columns=['daily_positive_benchmark', 'daily_negative_benchmark']) + df = df.drop(columns=["daily_positive_benchmark", "daily_negative_benchmark"]) print("日级别偏离度计算完成 (使用 pct_chg)。") return df @@ -2664,36 +2707,41 @@ def daily_industry_momentum_benchmark(df): print("--- 计算日级别行业动量基准 (使用 pct_chg 和 cat_l2_code) ---") # 确保必需列存在 - required_cols = ['pct_chg', 'cat_l2_code', 'trade_date', 'ts_code'] + required_cols = ["pct_chg", "cat_l2_code", "trade_date", "ts_code"] if not all(col in df.columns for col in required_cols): print(f"错误: 计算日级别行业动量基准需要以下列: {required_cols}。") return df # 计算每日每个行业内部的动量基准 # 使用 trade_date 和 cat_l2_code 进行分组 - industry_daily_benchmarks = df.groupby(['trade_date', 'cat_l2_code'])['pct_chg'].agg( - daily_industry_positive_benchmark = lambda x: x[x > 0].mean(), # 日级别行业内上涨股票的平均涨跌幅 - daily_industry_negative_benchmark = lambda x: x[x < 0].mean() # 日级别行业内下跌股票的平均涨跌幅 - ).reset_index() + industry_daily_benchmarks = ( + df.groupby(["trade_date", "cat_l2_code"])["pct_chg"] + .agg( + daily_industry_positive_benchmark=lambda x: x[ + x > 0 + ].mean(), # 日级别行业内上涨股票的平均涨跌幅 + daily_industry_negative_benchmark=lambda x: x[ + x < 0 + ].mean(), # 日级别行业内下跌股票的平均涨跌幅 + ) + .reset_index() + ) # 将日级别行业动量基准合并回原始日线数据DataFrame # 使用 trade_date 和 cat_l2_code 进行 merge df = pd.merge( - df, - industry_daily_benchmarks, - on=['trade_date', 'cat_l2_code'], - how='left' + df, industry_daily_benchmarks, on=["trade_date", "cat_l2_code"], how="left" ) # 对可能出现的NaN基准进行填充（例如某个行业某一天没有上涨或下跌的股票） # 这里用0填充表示该行业该天没有对应的同向基准 - df['daily_industry_positive_benchmark'].fillna(0, inplace=True) - df['daily_industry_negative_benchmark'].fillna(0, inplace=True) - + df["daily_industry_positive_benchmark"].fillna(0, inplace=True) + df["daily_industry_negative_benchmark"].fillna(0, inplace=True) print("日级别行业动量基准计算完成 (使用 pct_chg 和 cat_l2_code)。") return df + def daily_industry_deviation(df): """ 计算日级别行业偏离度，使用现有的 'pct_chg' 列和计算出的日级别行业动量基准。 @@ -2713,9 +2761,15 @@ def daily_industry_deviation(df): # 确保所需的列存在 df = daily_industry_momentum_benchmark(df) - required_cols = ['pct_chg', 'daily_industry_positive_benchmark', 'daily_industry_negative_benchmark'] + required_cols = [ + "pct_chg", + "daily_industry_positive_benchmark", + "daily_industry_negative_benchmark", + ] if not all(col in df.columns for col in required_cols): - print(f"错误: 计算日级别行业偏离度需要以下列: {required_cols}。请先运行 daily_industry_momentum_benchmark(df)。") + print( + f"错误: 计算日级别行业偏离度需要以下列: {required_cols}。请先运行 daily_industry_momentum_benchmark(df)。" + ) return df # 根据规则计算日级别行业偏离度： @@ -2724,61 +2778,86 @@ def daily_industry_deviation(df): # 否则 deviation = 0 conditions = [ - (df['pct_chg'] > 0) & (df['daily_industry_positive_benchmark'] > 0), - (df['pct_chg'] < 0) & (df['daily_industry_negative_benchmark'] < 0) + (df["pct_chg"] > 0) & (df["daily_industry_positive_benchmark"] > 0), + (df["pct_chg"] < 0) & (df["daily_industry_negative_benchmark"] < 0), ] choices = [ - df['pct_chg'] - df['daily_industry_positive_benchmark'], - df['pct_chg'] - df['daily_industry_negative_benchmark'] + df["pct_chg"] - df["daily_industry_positive_benchmark"], + df["pct_chg"] - df["daily_industry_negative_benchmark"], ] - df['daily_industry_deviation'] = np.select(conditions, choices, default=0) + df["daily_industry_deviation"] = np.select(conditions, choices, default=0) - df = df.drop(columns=['daily_industry_positive_benchmark', 'daily_industry_negative_benchmark']) + df = df.drop( + columns=[ + "daily_industry_positive_benchmark", + "daily_industry_negative_benchmark", + ] + ) print("日级别行业偏离度计算完成 (使用 pct_chg 和行业基准)。") return df -def sentiment_panic_greed_index(df: pd.DataFrame, window_atr: int = 14, window_smooth: int = 5, factor_name: str = 'senti_panic_greed'): +def sentiment_panic_greed_index( + df: pd.DataFrame, + window_atr: int = 14, + window_smooth: int = 5, + factor_name: str = "senti_panic_greed", +): """ 计算市场恐慌/贪婪指数 (原地修改)。 结合日内振幅、影线、跳空及与近期ATR的比较。 WARNING: Modifies df in-place. """ print(f"Calculating {factor_name}...") - _temp_cols = ['_prev_close', '_atr', '_true_range', '_upper_shadow', '_lower_shadow', '_body', '_gap', '_volatility_surprise'] - if not all(col in df.columns for col in ['open', 'high', 'low', 'close', 'vol']): + _temp_cols = [ + "_prev_close", + "_atr", + "_true_range", + "_upper_shadow", + "_lower_shadow", + "_body", + "_gap", + "_volatility_surprise", + ] + if not all(col in df.columns for col in ["open", "high", "low", "close", "vol"]): print(f"Error: DataFrame 缺少必需的 OHLCV 列。将为 {factor_name} 填充 NaN。") df[factor_name] = np.nan return - + try: - df['_prev_close'] = df['close'].shift(1) + df["_prev_close"] = df["close"].shift(1) # 计算真实波幅 (TR) 和 ATR - df['_true_range'] = talib.TRANGE(df['high'], df['low'], df['_prev_close']) - df['_atr'] = talib.ATR(df['high'], df['low'], df['_prev_close'], timeperiod=window_atr) + df["_true_range"] = talib.TRANGE(df["high"], df["low"], df["_prev_close"]) + df["_atr"] = talib.ATR( + df["high"], df["low"], df["_prev_close"], timeperiod=window_atr + ) # 计算影线和实体 - df['_upper_shadow'] = df['high'] - np.maximum(df['open'], df['close']) - df['_lower_shadow'] = np.minimum(df['open'], df['close']) - df['low'] - df['_body'] = np.abs(df['close'] - df['open']) + df["_upper_shadow"] = df["high"] - np.maximum(df["open"], df["close"]) + df["_lower_shadow"] = np.minimum(df["open"], df["close"]) - df["low"] + df["_body"] = np.abs(df["close"] - df["open"]) # 计算跳空 - df['_gap'] = (df['open'] / df['_prev_close'] - 1).fillna(0) + df["_gap"] = (df["open"] / df["_prev_close"] - 1).fillna(0) # 波动性意外: 当日真实波幅相对于近期ATR的倍数，乘以涨跌方向 # 如果真实波幅显著放大，根据涨跌幅赋予正负号，表明情绪的强度和方向 - df['_volatility_surprise'] = (df['_true_range'] / (df['_atr'] + epsilon) -1) * np.sign(df['pct_chg'].fillna(0)) - + df["_volatility_surprise"] = ( + df["_true_range"] / (df["_atr"] + epsilon) - 1 + ) * np.sign(df["pct_chg"].fillna(0)) # 简化版情绪指标：(下影线 - 上影线) / ATR + 跳空幅度 + 当日涨跌幅, 然后平滑 # 更强的信号：波动性意外，结合跳空 # 考虑当日振幅相对于ATR的超额部分，并结合实体方向 # ( (真实波幅/ATR) * 涨跌方向 ) + 跳空幅度 - raw_senti = (df['_true_range'] / (df['_atr'] + epsilon)) * np.sign(df['pct_chg'].fillna(0)) + df['_gap'] * 2 # 放大跳空影响 + raw_senti = (df["_true_range"] / (df["_atr"] + epsilon)) * np.sign( + df["pct_chg"].fillna(0) + ) + df[ + "_gap" + ] * 2 # 放大跳空影响 df[factor_name] = raw_senti.rolling(window_smooth, min_periods=1).mean() - except Exception as e: print(f"Error calculating {factor_name}: {e}") df[factor_name] = np.nan @@ -2790,23 +2869,34 @@ def sentiment_panic_greed_index(df: pd.DataFrame, window_atr: int = 14, window_s return df -def sentiment_market_breadth_proxy(df: pd.DataFrame, window_vol: int = 20, window_smooth: int = 3, factor_name: str = 'senti_breadth_proxy'): +def sentiment_market_breadth_proxy( + df: pd.DataFrame, + window_vol: int = 20, + window_smooth: int = 3, + factor_name: str = "senti_breadth_proxy", +): """ 计算市场宽度情绪代理指标 (基于指数的价量配合度) (原地修改). WARNING: Modifies df in-place. """ print(f"Calculating {factor_name}...") - _temp_cols = ['_rolling_avg_vol'] - if not all(col in df.columns for col in ['pct_chg', 'vol']): - print(f"Error: DataFrame 缺少 'pct_chg' 或 'vol' 列。将为 {factor_name} 填充 NaN。") + _temp_cols = ["_rolling_avg_vol"] + if not all(col in df.columns for col in ["pct_chg", "vol"]): + print( + f"Error: DataFrame 缺少 'pct_chg' 或 'vol' 列。将为 {factor_name} 填充 NaN。" + ) df[factor_name] = np.nan return try: - df['_rolling_avg_vol'] = df['vol'].rolling(window_vol, min_periods=max(1, window_vol//2)).mean() + df["_rolling_avg_vol"] = ( + df["vol"].rolling(window_vol, min_periods=max(1, window_vol // 2)).mean() + ) # 价量配合度：涨跌幅乘以相对成交量强度 - raw_breadth = df['pct_chg'] * (df['vol'] / (df['_rolling_avg_vol'] + epsilon)) - df[factor_name] = raw_breadth.rolling(window_smooth, min_periods=1).mean() # 平滑处理 + raw_breadth = df["pct_chg"] * (df["vol"] / (df["_rolling_avg_vol"] + epsilon)) + df[factor_name] = raw_breadth.rolling( + window_smooth, min_periods=1 + ).mean() # 平滑处理 except Exception as e: print(f"Error calculating {factor_name}: {e}") @@ -2817,29 +2907,37 @@ def sentiment_market_breadth_proxy(df: pd.DataFrame, window_vol: int = 20, windo df.drop(columns=cols_to_drop, inplace=True) print(f"Finished {factor_name}.") return df - -def sentiment_reversal_indicator(df: pd.DataFrame, window_ret: int = 5, window_vol: int = 5, factor_name: str = 'senti_reversal'): + + +def sentiment_reversal_indicator( + df: pd.DataFrame, + window_ret: int = 5, + window_vol: int = 5, + factor_name: str = "senti_reversal", +): """ 计算短期情绪反转因子 (原地修改). WARNING: Modifies df in-place. """ print(f"Calculating {factor_name}...") - _temp_cols = ['_return_M', '_volatility_M'] - if 'pct_chg' not in df.columns: + _temp_cols = ["_return_M", "_volatility_M"] + if "pct_chg" not in df.columns: print(f"Error: DataFrame 缺少 'pct_chg' 列。将为 {factor_name} 填充 NaN。") df[factor_name] = np.nan return try: # 计算 M 日累计收益率 (这里用连乘近似，或者 sum of log returns) # (close / close.shift(M)) -1 - df['_return_M'] = (df['close'] / df['close'].shift(window_ret)) - 1 + df["_return_M"] = (df["close"] / df["close"].shift(window_ret)) - 1 # df['_return_M'] = df['pct_chg'].rolling(window_ret, min_periods=1).sum() # 另一种近似 # 计算 M 日已实现波动率 - df['_volatility_M'] = df['pct_chg'].rolling(window_vol, min_periods=max(1, window_vol//2)).std() + df["_volatility_M"] = ( + df["pct_chg"].rolling(window_vol, min_periods=max(1, window_vol // 2)).std() + ) # 因子计算 - df[factor_name] = -df['_return_M'] * df['_volatility_M'] + df[factor_name] = -df["_return_M"] * df["_volatility_M"] # 对因子本身可以再做一次平滑 # df[factor_name] = df[factor_name].rolling(3, min_periods=1).mean() @@ -2868,28 +2966,29 @@ def price_minus_deduction_price(df, n=10): 返回: pd.DataFrame: 增加了 'price_minus_deduction_price_n' 列的DataFrame。 """ - if 'close' not in df.columns: + if "close" not in df.columns: print("错误: DataFrame中没有'close'列，无法计算抵扣价相关因子。") return df if n <= 1: print("错误: 均线周期 n 必须大于 1。") - df[f'price_minus_deduction_price_{n}'] = np.nan + df[f"price_minus_deduction_price_{n}"] = np.nan return df - grouped = df.groupby('ts_code', group_keys=False) + grouped = df.groupby("ts_code", group_keys=False) # 抵扣价是当前窗口移除的最早的价格，即当前价格的 n-1 周期前的价格 # 例如计算 SMA(10) for P_t, 窗口是 P_{t-9}, ..., P_t. 移除的是 P_{t-9}. # P_{t-9} 是 P_t 的 shift(9). So shift(n-1). - deduction_price = grouped['close'].shift(n - 1) + deduction_price = grouped["close"].shift(n - 1) # 计算差值 - df[f'price_minus_deduction_price_{n}'] = df['close'] - deduction_price + df[f"price_minus_deduction_price_{n}"] = df["close"] - deduction_price print(f"因子 price_minus_deduction_price_{n} 计算完成。") return df + def price_deduction_price_diff_ratio_to_sma(df, n=10): """ 因子 2 (定量): 计算当前收盘价与抵扣价的差值占N周期均线的比例。 @@ -2903,32 +3002,32 @@ def price_deduction_price_diff_ratio_to_sma(df, n=10): 返回: pd.DataFrame: 增加了 'price_deduction_price_diff_ratio_to_sma_n' 列的DataFrame。 """ - if 'close' not in df.columns: + if "close" not in df.columns: print("错误: DataFrame中没有'close'列，无法计算抵扣价相关因子。") return df if n <= 1: print("错误: 均线周期 n 必须大于 1。") - df[f'price_deduction_price_diff_ratio_to_sma_{n}'] = np.nan + df[f"price_deduction_price_diff_ratio_to_sma_{n}"] = np.nan return df - grouped = df.groupby('ts_code', group_keys=False) + grouped = df.groupby("ts_code", group_keys=False) # 计算N周期SMA # 使用 transform 可以保持与原始 df 的索引对齐 - sma = grouped['close'].transform(lambda x: x.rolling(window=n).mean()) + sma = grouped["close"].transform(lambda x: x.rolling(window=n).mean()) # 抵扣价 - deduction_price = grouped['close'].shift(n - 1) + deduction_price = grouped["close"].shift(n - 1) # 计算比例，处理均线为零的情况 - diff = df['close'] - deduction_price + diff = df["close"] - deduction_price # 使用 np.divide 并指定 where 条件和 fill_value 来避免除以零警告和 NaN 结果 # 如果 sma 为 0，则结果设为 NaN - df[f'price_deduction_price_diff_ratio_to_sma_{n}'] = np.divide( + df[f"price_deduction_price_diff_ratio_to_sma_{n}"] = np.divide( diff, sma, - out=np.full_like(diff, np.nan), # 输出数组形状与 diff 相同，NaN 填充 - where=(sma != 0) # 仅在 sma 不为 0 时执行除法 + out=np.full_like(diff, np.nan), # 输出数组形状与 diff 相同，NaN 填充 + where=(sma != 0), # 仅在 sma 不为 0 时执行除法 ) # np.divide with where handles Inf/-Inf and 0/0 (as NaN), but explicitly replace might be slightly safer depending on numpy version @@ -2958,46 +3057,51 @@ def cat_price_vs_sma_vs_deduction_price(df, n=10): 返回: pd.DataFrame: 增加了 'cat_price_vs_sma_vs_deduction_price_n' 列的DataFrame。 """ - if 'close' not in df.columns: + if "close" not in df.columns: print("错误: DataFrame中没有'close'列，无法计算抵扣价相关因子。") return df if n <= 1: print("错误: 均线周期 n 必须大于 1。") - df[f'cat_price_vs_sma_vs_deduction_price_{n}'] = np.nan + df[f"cat_price_vs_sma_vs_deduction_price_{n}"] = np.nan return df - grouped = df.groupby('ts_code', group_keys=False) + grouped = df.groupby("ts_code", group_keys=False) # 计算N周期SMA - sma = grouped['close'].transform(lambda x: x.rolling(window=n).mean()) + sma = grouped["close"].transform(lambda x: x.rolling(window=n).mean()) # 抵扣价 - deduction_price = grouped['close'].shift(n - 1) + deduction_price = grouped["close"].shift(n - 1) # 定义条件和分类值 conditions = [ - (df['close'] > sma) & (deduction_price > sma), - (df['close'] < sma) & (deduction_price < sma), - (df['close'] > sma) & (deduction_price <= sma), # 包含等于的情况 - (df['close'] <= sma) & (deduction_price > sma) # 包含等于的情况 + (df["close"] > sma) & (deduction_price > sma), + (df["close"] < sma) & (deduction_price < sma), + (df["close"] > sma) & (deduction_price <= sma), # 包含等于的情况 + (df["close"] <= sma) & (deduction_price > sma), # 包含等于的情况 # 注意：sma 或 deduction_price 为 NaN 的行，其条件结果为 False，会落入 default=0 ] choices = [1, 2, 3, 4] # 使用 np.select 进行分类 # 默认值为 0，用于处理条件不满足或输入为 NaN 的情况 - df[f'cat_price_vs_sma_vs_deduction_price_{n}'] = np.select(conditions, choices, default=0) + df[f"cat_price_vs_sma_vs_deduction_price_{n}"] = np.select( + conditions, choices, default=0 + ) print(f"因子 cat_price_vs_sma_vs_deduction_price_{n} 计算完成。") return df + def cat_is_on_top_list(df: pd.DataFrame, top_list: pd.DataFrame) -> pd.DataFrame: - if 'cat_reason' not in df.columns: - print('计算因子cat_is_on_top_list失败，缺少cat_reason列') + if "cat_reason" not in df.columns: + print("计算因子cat_is_on_top_list失败，缺少cat_reason列") return df - df['cat_is_on_top_list'] = df['cat_reason'] - df['cat_is_on_top_list'] = df['cat_is_on_top_list'] * (df['pct_chg'] > 0).astype(int) + df["cat_is_on_top_list"] = df["cat_reason"] + df["cat_is_on_top_list"] = df["cat_is_on_top_list"] * (df["pct_chg"] > 0).astype( + int + ) return df @@ -3014,23 +3118,238 @@ def cat_reason(df: pd.DataFrame, top_list: pd.DataFrame) -> pd.DataFrame: pd.DataFrame: 包含 'ts_code', 'trade_date' 和 'cat_reason' 列。 """ # 提取所有唯一的 reason 并进行数值编码 - unique_reasons = top_list['reason'].unique() + unique_reasons = top_list["reason"].unique() reason_mapping = {reason: i + 1 for i, reason in enumerate(unique_reasons)} # 在 top_list 中创建数值型的 reason 列 - top_list['cat_reason'] = top_list['reason'].map(reason_mapping).astype('Int64') + top_list["cat_reason"] = top_list["reason"].map(reason_mapping).astype("Int64") # 转换 trade_date 类型以进行合并 - df['trade_date'] = pd.to_datetime(df['trade_date']) - top_list['trade_date'] = pd.to_datetime(top_list['trade_date']) + df["trade_date"] = pd.to_datetime(df["trade_date"]) + top_list["trade_date"] = pd.to_datetime(top_list["trade_date"]) # 仅保留 top_list 中需要的列进行合并 - top_list_slim = top_list[['ts_code', 'trade_date', 'cat_reason']] + top_list_slim = top_list[["ts_code", "trade_date", "cat_reason"]] # 合并 DataFrame - merged_df = df.merge(top_list_slim, on=['ts_code', 'trade_date'], how='left') + merged_df = df.merge(top_list_slim, on=["ts_code", "trade_date"], how="left") # 填充 NaN 为 0 并转换为 int 类型 - merged_df['cat_reason'] = merged_df['cat_reason'].fillna(0).astype('int') + merged_df["cat_reason"] = merged_df["cat_reason"].fillna(0).astype("int") - return merged_df \ No newline at end of file + return merged_df + + +def ts_volatility_slope_20_5(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算 20 日收益率标准差的 5 日线性回归斜率因子。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'pct_chg' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_volatility_slope_20_5' 列的 DataFrame。 + """ + df["trade_date"] = pd.to_datetime(df["trade_date"]) + df.sort_values(["ts_code", "trade_date"], inplace=True) + + def std_slope(series): + if len(series) < 2: + return 0 + x = np.arange(len(series)) + slope, _, _, _, _ = linregress(x, series) + return slope + + df["volatility_20"] = ( + df.groupby("ts_code")["pct_chg"] + .rolling(window=20, min_periods=1) + .std() + .reset_index(level=0, drop=True) + ) + df["ts_volatility_slope_20_5"] = ( + df.groupby("ts_code")["volatility_20"] + .rolling(window=5, min_periods=2) + .apply(std_slope) + .reset_index(level=0, drop=True) + ) + df.drop(columns=["volatility_20"], inplace=True) + return df + + +def ts_turnover_rate_acceleration_5_20(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算短期 (5日) 和长期 (20日) 换手率均值的差值因子。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'turnover_rate' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_turnover_rate_acceleration_5_20' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + df['short_avg_turnover'] = df.groupby('ts_code')['turnover_rate'].rolling(window=5, min_periods=1).mean().reset_index(level=0, drop=True) + df['long_avg_turnover'] = df.groupby('ts_code')['turnover_rate'].rolling(window=20, min_periods=1).mean().reset_index(level=0, drop=True) + df['ts_turnover_rate_acceleration_5_20'] = df['short_avg_turnover'] - df['long_avg_turnover'] + df.drop(columns=['short_avg_turnover', 'long_avg_turnover'], inplace=True) + return df + +def ts_vol_sustain_10_30(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算过去 10 日成交量大于 30 日均值的交易天数占比因子。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'vol' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_vol_sustain_10_30' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + df['long_avg_vol'] = df.groupby('ts_code')['vol'].rolling(window=30, min_periods=1).mean().reset_index(level=0, drop=True) + + def vol_above_avg(group): + group['vol_above'] = group['vol'] > group['long_avg_vol'] + group['ts_vol_sustain_10_30'] = group['vol_above'].rolling(window=10, min_periods=1).mean() + return group.drop(columns=['vol_above']) + + df = df.groupby('ts_code', group_keys=False).apply(vol_above_avg) + df.drop(columns=['long_avg_vol'], inplace=True) + return df + +def cs_turnover_rate_relative_strength_20(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算当日换手率 vs 20 日均值比值的横截面排名因子。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'turnover_rate' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'cs_turnover_rate_relative_strength_20' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + + def calculate_ratio(group): + group['avg_turnover_20'] = group['turnover_rate'].rolling(window=20, min_periods=1).mean() + group['turnover_ratio'] = group['turnover_rate'] / group['avg_turnover_20'] + return group.drop(columns=['avg_turnover_20']) + + df = df.groupby('ts_code', group_keys=False).apply(calculate_ratio) + + def rank_ratios(group): + group['cs_turnover_rate_relative_strength_20'] = group['turnover_ratio'].rank(method='dense', ascending=False) + return group.drop(columns=['turnover_ratio']) + + df = df.groupby('trade_date', group_keys=False).apply(rank_ratios) + return df + +def cs_amount_outlier_10(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算当日成交额 vs 10 日均值差值的横截面 Z-score 因子。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'amount' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'cs_amount_outlier_10' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + + def calculate_diff(group): + group['avg_amount_10'] = group['amount'].rolling(window=10, min_periods=1).mean() + group['amount_diff'] = group['amount'] - group['avg_amount_10'] + return group.drop(columns=['avg_amount_10']) + + df = df.groupby('ts_code', group_keys=False).apply(calculate_diff) + + def zscore_diff(group): + mean_diff = group['amount_diff'].mean() + std_diff = group['amount_diff'].std() + if std_diff == 0: + group['cs_amount_outlier_10'] = 0 + else: + group['cs_amount_outlier_10'] = (group['amount_diff'] - mean_diff) / std_diff + return group.drop(columns=['amount_diff']) + + df = df.groupby('trade_date', group_keys=False).apply(zscore_diff) + return df + +def ts_ff_to_total_turnover_ratio(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算自由流通股换手率与总换手率之比。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date', 'turnover_rate' 和 'turnover_rate_ff' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_ff_to_total_turnover_ratio' 列的 DataFrame。 + """ + df['ts_ff_to_total_turnover_ratio'] = df['turnover_rate_ff'] / (df['turnover_rate'] + 1e-8) # 避免除零 + return df + + +def ts_price_volume_trend_coherence_5_20(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算过去 5 日价格上涨占比与过去 5 日成交量高于 20 日均量占比的乘积。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date', 'close' 和 'vol' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_price_volume_trend_coherence_5_20' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + + def price_up_days(series): + return (series.diff() > 0).rolling(window=5, min_periods=1).mean() + + df['price_up_ratio'] = df.groupby('ts_code')['close'].apply(price_up_days).reset_index(level=0, drop=True) + df['vol_avg_20'] = df.groupby('ts_code')['vol'].rolling(window=20, min_periods=1).mean().reset_index(level=0, drop=True) + df['vol_above_avg'] = (df['vol'] > df['vol_avg_20']).rolling(window=5, min_periods=1).mean() + df['ts_price_volume_trend_coherence_5_20'] = df['price_up_ratio'] * df['vol_above_avg'] + df.drop(columns=['price_up_ratio', 'vol_avg_20', 'vol_above_avg'], inplace=True) + return df + +def ts_turnover_rate_trend_strength_5(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算过去 5 日换手率的线性回归斜率。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'turnover_rate' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_turnover_rate_trend_strength_5' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + + def turnover_slope(series): + if len(series) < 2: + return 0 + x = np.arange(len(series)) + slope, _, _, _, _ = linregress(x, series) + return slope + + df['ts_turnover_rate_trend_strength_5'] = df.groupby('ts_code')['turnover_rate'].rolling(window=5, min_periods=2).apply(turnover_slope).reset_index(level=0, drop=True) + return df + + +def ts_ff_turnover_rate_surge_10(df: pd.DataFrame) -> pd.DataFrame: + """ + 计算当日自由流通股换手率与过去 10 日均值比值。 + + Args: + df (pd.DataFrame): 包含 'ts_code', 'trade_date' 和 'turnover_rate_ff' 列的 DataFrame。 + + Returns: + pd.DataFrame: 包含新增 'ts_ff_turnover_rate_surge_10' 列的 DataFrame。 + """ + df['trade_date'] = pd.to_datetime(df['trade_date']) + df.sort_values(['ts_code', 'trade_date'], inplace=True) + df['avg_ff_turnover_10'] = df.groupby('ts_code')['turnover_rate_ff'].rolling(window=10, min_periods=1).mean().reset_index(level=0, drop=True) + df['ts_ff_turnover_rate_surge_10'] = df['turnover_rate_ff'] / (df['avg_ff_turnover_10'] + 1e-8) # 避免除零 + df.drop(columns=['avg_ff_turnover_10'], inplace=True) + return df diff --git a/main/train/Classify2.ipynb b/main/train/Classify2.ipynb index 18c68ad..aa6e54b 100644 --- a/main/train/Classify2.ipynb +++ b/main/train/Classify2.ipynb @@ -39,12 +39,24 @@ "\n", "import warnings\n", "\n", - "warnings.filterwarnings(\"ignore\")" + "warnings.filterwarnings(\"ignore\")\n" ] }, { "cell_type": "code", "execution_count": 2, + "id": "4a481c60", + "metadata": {}, + "outputs": [], + "source": [ + "# 设置使用核心\n", + "import os\n", + "os.environ[\"MODIN_CPUS\"] = \"4\"\n" + ] + }, + { + "cell_type": "code", + "execution_count": 3, "id": "a79cafb06a7e0e43", "metadata": { "ExecuteTime": { @@ -58,13 +70,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "daily data\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ + "daily data\n", "daily basic\n", "inner merge on ['ts_code', 'trade_date']\n", "stk limit\n", @@ -75,7 +81,7 @@ "left merge on ['ts_code', 'trade_date']\n", "\n", "RangeIndex: 8665405 entries, 0 to 8665404\n", - "Data columns (total 32 columns):\n", + "Data columns (total 33 columns):\n", " # Column Dtype \n", "--- ------ ----- \n", " 0 ts_code object \n", @@ -85,33 +91,34 @@ " 4 high float64 \n", " 5 low float64 \n", " 6 vol float64 \n", - " 7 pct_chg float64 \n", - " 8 turnover_rate float64 \n", - " 9 pe_ttm float64 \n", - " 10 circ_mv float64 \n", - " 11 total_mv float64 \n", - " 12 volume_ratio float64 \n", - " 13 is_st bool \n", - " 14 up_limit float64 \n", - " 15 down_limit float64 \n", - " 16 buy_sm_vol float64 \n", - " 17 sell_sm_vol float64 \n", - " 18 buy_lg_vol float64 \n", - " 19 sell_lg_vol float64 \n", - " 20 buy_elg_vol float64 \n", - " 21 sell_elg_vol float64 \n", - " 22 net_mf_vol float64 \n", - " 23 his_low float64 \n", - " 24 his_high float64 \n", - " 25 cost_5pct float64 \n", - " 26 cost_15pct float64 \n", - " 27 cost_50pct float64 \n", - " 28 cost_85pct float64 \n", - " 29 cost_95pct float64 \n", - " 30 weight_avg float64 \n", - " 31 winner_rate float64 \n", - "dtypes: bool(1), datetime64[ns](1), float64(29), object(1)\n", - "memory usage: 2.0+ GB\n", + " 7 amount float64 \n", + " 8 pct_chg float64 \n", + " 9 turnover_rate float64 \n", + " 10 pe_ttm float64 \n", + " 11 circ_mv float64 \n", + " 12 total_mv float64 \n", + " 13 volume_ratio float64 \n", + " 14 is_st bool \n", + " 15 up_limit float64 \n", + " 16 down_limit float64 \n", + " 17 buy_sm_vol float64 \n", + " 18 sell_sm_vol float64 \n", + " 19 buy_lg_vol float64 \n", + " 20 sell_lg_vol float64 \n", + " 21 buy_elg_vol float64 \n", + " 22 sell_elg_vol float64 \n", + " 23 net_mf_vol float64 \n", + " 24 his_low float64 \n", + " 25 his_high float64 \n", + " 26 cost_5pct float64 \n", + " 27 cost_15pct float64 \n", + " 28 cost_50pct float64 \n", + " 29 cost_85pct float64 \n", + " 30 cost_95pct float64 \n", + " 31 weight_avg float64 \n", + " 32 winner_rate float64 \n", + "dtypes: bool(1), datetime64[ns](1), float64(30), object(1)\n", + "memory usage: 2.1+ GB\n", "None\n" ] } @@ -121,7 +128,7 @@ "\n", "print('daily data')\n", "df = read_and_merge_h5_data('../../data/daily_data.h5', key='daily_data',\n", - " columns=['ts_code', 'trade_date', 'open', 'close', 'high', 'low', 'vol', 'pct_chg'],\n", + " columns=['ts_code', 'trade_date', 'open', 'close', 'high', 'low', 'vol', 'amount', 'pct_chg'],\n", " df=None)\n", "\n", "print('daily basic')\n", @@ -149,7 +156,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 4, "id": "cac01788dac10678", "metadata": { "ExecuteTime": { @@ -217,7 +224,7 @@ }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 5, "id": "c4e9e1d31da6dba6", "metadata": { "ExecuteTime": { @@ -317,7 +324,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 6, "id": "a735bc02ceb4d872", "metadata": { "ExecuteTime": { @@ -333,7 +340,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 7, "id": "53f86ddc0677a6d7", "metadata": { "ExecuteTime": { @@ -400,7 +407,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 8, "id": "dbe2fd8021b9417f", "metadata": { "ExecuteTime": { @@ -413,7 +420,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "['ts_code', 'open', 'close', 'high', 'low', 'circ_mv', 'total_mv', 'is_st', 'up_limit', 'down_limit', 'buy_sm_vol', 'sell_sm_vol', 'buy_lg_vol', 'sell_lg_vol', 'buy_elg_vol', 'sell_elg_vol', 'net_mf_vol', 'his_low', 'his_high', 'cost_5pct', 'cost_15pct', 'cost_50pct', 'cost_85pct', 'cost_95pct', 'weight_avg', 'in_date']\n" + "['ts_code', 'open', 'close', 'high', 'low', 'amount', 'circ_mv', 'total_mv', 'is_st', 'up_limit', 'down_limit', 'buy_sm_vol', 'sell_sm_vol', 'buy_lg_vol', 'sell_lg_vol', 'buy_elg_vol', 'sell_elg_vol', 'net_mf_vol', 'his_low', 'his_high', 'cost_5pct', 'cost_15pct', 'cost_50pct', 'cost_85pct', 'cost_95pct', 'weight_avg', 'in_date']\n" ] } ], @@ -428,7 +435,7 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": 9, "id": "85c3e3d0235ffffa", "metadata": { "ExecuteTime": { @@ -456,7 +463,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": 10, "id": "92d84ce15a562ec6", "metadata": { "ExecuteTime": { @@ -476,7 +483,13 @@ "使用 'ann_date' 作为财务数据生效日期。\n", "警告: 从 financial_data_subset 中移除了 366 行，因为其 'ts_code' 或 'ann_date' 列存在空值。\n", "使用 'ann_date' 作为财务数据生效日期。\n", - "警告: 从 financial_data_subset 中移除了 366 行，因为其 'ts_code' 或 'ann_date' 列存在空值。\n", + "警告: 从 financial_data_subset 中移除了 366 行，因为其 'ts_code' 或 'ann_date' 列存在空值。\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ "开始计算因子: AR, BR (原地修改)...\n", "因子 AR, BR 计算成功。\n", "因子 AR, BR 计算流程结束。\n", @@ -495,7 +508,7 @@ "错误: 计算日级别行业动量基准需要以下列: ['pct_chg', 'cat_l2_code', 'trade_date', 'ts_code']。\n", "错误: 计算日级别行业偏离度需要以下列: ['pct_chg', 'daily_industry_positive_benchmark', 'daily_industry_negative_benchmark']。请先运行 daily_industry_momentum_benchmark(df)。\n", "Index(['ts_code', 'trade_date', 'open', 'close', 'high', 'low', 'vol',\n", - " 'pct_chg', 'turnover_rate', 'pe_ttm', 'circ_mv', 'total_mv',\n", + " 'amount', 'pct_chg', 'turnover_rate', 'pe_ttm', 'circ_mv', 'total_mv',\n", " 'volume_ratio', 'is_st', 'up_limit', 'down_limit', 'buy_sm_vol',\n", " 'sell_sm_vol', 'buy_lg_vol', 'sell_lg_vol', 'buy_elg_vol',\n", " 'sell_elg_vol', 'net_mf_vol', 'his_low', 'his_high', 'cost_5pct',\n", @@ -592,11 +605,11 @@ "Finished cs_rank_size.\n", "\n", "RangeIndex: 4539678 entries, 0 to 4539677\n", - "Columns: 180 entries, ts_code to cs_rank_size\n", - "dtypes: bool(10), datetime64[ns](1), float64(164), int32(3), object(2)\n", - "memory usage: 5.7+ GB\n", + "Columns: 181 entries, ts_code to cs_rank_size\n", + "dtypes: bool(10), datetime64[ns](1), float64(165), int32(3), object(2)\n", + "memory usage: 5.8+ GB\n", "None\n", - "['ts_code', 'trade_date', 'open', 'close', 'high', 'low', 'vol', 'pct_chg', 'turnover_rate', 'pe_ttm', 'circ_mv', 'total_mv', 'volume_ratio', 'is_st', 'up_limit', 'down_limit', 'buy_sm_vol', 'sell_sm_vol', 'buy_lg_vol', 'sell_lg_vol', 'buy_elg_vol', 'sell_elg_vol', 'net_mf_vol', 'his_low', 'his_high', 'cost_5pct', 'cost_15pct', 'cost_50pct', 'cost_85pct', 'cost_95pct', 'weight_avg', 'winner_rate', 'cat_l2_code', 'undist_profit_ps', 'ocfps', 'roa', 'roe', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'flow_divergence_diff', 'flow_divergence_ratio', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'price_cost_divergence', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_flow_divergence', 'cs_rank_ind_adj_lg_flow', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_opening_gap', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_ind_cap_neutral_pe', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size']\n" + "['ts_code', 'trade_date', 'open', 'close', 'high', 'low', 'vol', 'amount', 'pct_chg', 'turnover_rate', 'pe_ttm', 'circ_mv', 'total_mv', 'volume_ratio', 'is_st', 'up_limit', 'down_limit', 'buy_sm_vol', 'sell_sm_vol', 'buy_lg_vol', 'sell_lg_vol', 'buy_elg_vol', 'sell_elg_vol', 'net_mf_vol', 'his_low', 'his_high', 'cost_5pct', 'cost_15pct', 'cost_50pct', 'cost_85pct', 'cost_95pct', 'weight_avg', 'winner_rate', 'cat_l2_code', 'undist_profit_ps', 'ocfps', 'roa', 'roe', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'flow_divergence_diff', 'flow_divergence_ratio', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'price_cost_divergence', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_flow_divergence', 'cs_rank_ind_adj_lg_flow', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_opening_gap', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_ind_cap_neutral_pe', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size']\n" ] } ], @@ -629,6 +642,11 @@ "# df = cat_reason(df, top_list_df)\n", "# df = cat_is_on_top_list(df, top_list_df)\n", "\n", + "# df = ts_turnover_rate_acceleration_5_20(df)\n", + "# df = ts_vol_sustain_10_30(df)\n", + "# df = cs_turnover_rate_relative_strength_20(df)\n", + "# df = cs_amount_outlier_10(df)\n", + "\n", "df = add_financial_factor(df, fina_indicator_df, factor_value_col='undist_profit_ps')\n", "df = add_financial_factor(df, fina_indicator_df, factor_value_col='ocfps')\n", "df = add_financial_factor(df, fina_indicator_df, factor_value_col='roa')\n", @@ -697,7 +715,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 11, "id": "b87b938028afa206", "metadata": { "ExecuteTime": { @@ -735,7 +753,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 12, "id": "f4f16d63ad18d1bc", "metadata": { "ExecuteTime": { @@ -745,7 +763,6 @@ }, "outputs": [], "source": [ - "import pandas as pd\n", "import numpy as np\n", "import statsmodels.api as sm # 用于中性化回归\n", "from tqdm import tqdm # 可选，用于显示进度条\n", @@ -962,7 +979,7 @@ }, { "cell_type": "code", - "execution_count": 12, + "execution_count": 13, "id": "40e6b68a91b30c79", "metadata": { "ExecuteTime": { @@ -972,9 +989,6 @@ }, "outputs": [], "source": [ - "import pandas as pd\n", - "\n", - "\n", "def remove_outliers_label_percentile(label: pd.Series, lower_percentile: float = 0.01, upper_percentile: float = 0.99,\n", " log=True):\n", " if not (0 <= lower_percentile < upper_percentile <= 1):\n", @@ -1071,10 +1085,6 @@ " return df_standardized\n", "\n", "\n", - "import numpy as np\n", - "import pandas as pd\n", - "\n", - "\n", "def neutralize_manual_revised(df: pd.DataFrame, features: list, industry_col: str, mkt_cap_col: str) -> pd.DataFrame:\n", " \"\"\"\n", " 手动实现简单回归以提升速度，通过构建 Series 确保索引对齐。\n", @@ -1289,7 +1299,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 14, "id": "47c12bb34062ae7a", "metadata": { "ExecuteTime": { @@ -1323,7 +1333,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 15, "id": "29221dde", "metadata": {}, "outputs": [ @@ -1366,7 +1376,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 16, "id": "03ee5daf", "metadata": {}, "outputs": [], @@ -1379,7 +1389,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 17, "id": "b76ea08a", "metadata": {}, "outputs": [ @@ -1400,7 +1410,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "MAD Filtering: 100%|██████████| 131/131 [00:28<00:00, 4.67it/s]\n" + "MAD Filtering: 100%|██████████| 131/131 [00:28<00:00, 4.59it/s]\n" ] }, { @@ -1415,7 +1425,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "MAD Filtering: 100%|██████████| 131/131 [00:24<00:00, 5.43it/s]\n" + "MAD Filtering: 100%|██████████| 131/131 [00:23<00:00, 5.52it/s]\n" ] }, { @@ -1597,7 +1607,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 18, "id": "3ff2d1c5", "metadata": {}, "outputs": [], @@ -1606,8 +1616,6 @@ "from sklearn.linear_model import LogisticRegression\n", "import matplotlib.pyplot as plt # 保持 matplotlib 导入，尽管LightGBM的绘图功能已移除\n", "from sklearn.decomposition import PCA\n", - "import pandas as pd\n", - "import numpy as np\n", "import datetime # 用于日期计算\n", "from catboost import CatBoostClassifier\n", "from catboost import Pool\n", @@ -1740,7 +1748,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 19, "id": "c6eb5cd4-e714-420a-ac48-39af3e11ee81", "metadata": { "ExecuteTime": { @@ -1774,7 +1782,7 @@ { "data": { "application/vnd.jupyter.widget-view+json": { - "model_id": "b132fdab733b4d4e856d4688f2997dac", + "model_id": "5119f1526efe46b58c2c2c978ebdf1f1", "version_major": 2, "version_minor": 0 }, @@ -1789,8 +1797,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "0:\tlearn: 0.6887138\ttest: 0.6894516\tbest: 0.6894516 (0)\ttotal: 328ms\tremaining: 8m 12s\n", - "bestTest = 0.5217666894\n", + "0:\tlearn: 0.6887139\ttest: 0.6894516\tbest: 0.6894516 (0)\ttotal: 271ms\tremaining: 6m 46s\n", + "bestTest = 0.5217666495\n", "bestIteration = 487\n", "Shrink model to first 488 iterations.\n" ] @@ -1814,7 +1822,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 20, "id": "5d1522a7538db91b", "metadata": { "ExecuteTime": { @@ -1852,7 +1860,7 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 21, "id": "09b1799e", "metadata": {}, "outputs": [ @@ -1874,7 +1882,7 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": 22, "id": "e53b209a", "metadata": {}, "outputs": [ @@ -1907,12 +1915,11 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 23, "id": "364e821a", "metadata": {}, "outputs": [], "source": [ - "import pandas as pd\n", "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, roc_auc_score, roc_curve\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", @@ -1992,7 +1999,7 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": 24, "id": "1f6e6336", "metadata": {}, "outputs": [ @@ -2061,7 +2068,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 25, "id": "7e9023cc", "metadata": {}, "outputs": [], @@ -2261,7 +2268,7 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 26, "id": "a0000d75", "metadata": {}, "outputs": [ @@ -2320,8 +2327,6 @@ } ], "source": [ - "import pandas as pd\n", - "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "from scipy.stats import spearmanr\n", @@ -2509,7 +2514,7 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": 27, "id": "a436dba4", "metadata": {}, "outputs": [ diff --git a/main/train/Rank2.ipynb b/main/train/Rank2.ipynb index cd8dbe6..f893ee5 100644 --- a/main/train/Rank2.ipynb +++ b/main/train/Rank2.ipynb @@ -450,7 +450,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": null, "id": "92d84ce15a562ec6", "metadata": { "ExecuteTime": { @@ -466,7 +466,13 @@ "使用 'ann_date' 作为财务数据生效日期。\n", "警告: 从 financial_data_subset 中移除了 366 行，因为其 'ts_code' 或 'ann_date' 列存在空值。\n", "使用 'ann_date' 作为财务数据生效日期。\n", - "警告: 从 financial_data_subset 中移除了 366 行，因为其 'ts_code' 或 'ann_date' 列存在空值。\n", + "警告: 从 financial_data_subset 中移除了 366 行，因为其 'ts_code' 或 'ann_date' 列存在空值。\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ "开始计算因子: AR, BR (原地修改)...\n", "因子 AR, BR 计算成功。\n", "因子 AR, BR 计算流程结束。\n", @@ -506,33 +512,86 @@ " 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike',\n", " 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike',\n", " 'vol_std_5', 'atr_14', 'atr_6', 'obv'],\n", - " dtype='object')\n" - ] - }, - { - "ename": "KeyboardInterrupt", - "evalue": "", - "output_type": "error", - "traceback": [ - "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[1;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)", - "Cell \u001b[1;32mIn[9], line 39\u001b[0m\n\u001b[0;32m 37\u001b[0m df \u001b[38;5;241m=\u001b[39m daily_deviation(df)\n\u001b[0;32m 38\u001b[0m df \u001b[38;5;241m=\u001b[39m daily_industry_deviation(df)\n\u001b[1;32m---> 39\u001b[0m df, _ \u001b[38;5;241m=\u001b[39m get_rolling_factor(df)\n\u001b[0;32m 40\u001b[0m df, _ \u001b[38;5;241m=\u001b[39m get_simple_factor(df)\n\u001b[0;32m 42\u001b[0m df \u001b[38;5;241m=\u001b[39m df\u001b[38;5;241m.\u001b[39mrename(columns\u001b[38;5;241m=\u001b[39m{\u001b[38;5;124m'\u001b[39m\u001b[38;5;124ml1_code\u001b[39m\u001b[38;5;124m'\u001b[39m: \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcat_l1_code\u001b[39m\u001b[38;5;124m'\u001b[39m})\n", - "File \u001b[1;32me:\\PyProject\\NewStock\\main\\train\\../..\\main\\factor\\factor.py:340\u001b[0m, in \u001b[0;36mget_rolling_factor\u001b[1;34m(df)\u001b[0m\n\u001b[0;32m 336\u001b[0m \u001b[38;5;66;03m# 计算 act_factor1, act_factor2, act_factor3, act_factor4\u001b[39;00m\n\u001b[0;32m 337\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor1\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_5\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 338\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m50\u001b[39m\n\u001b[0;32m 339\u001b[0m )\n\u001b[1;32m--> 340\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor2\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_13\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 341\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m40\u001b[39m\n\u001b[0;32m 342\u001b[0m )\n\u001b[0;32m 343\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor3\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_20\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 344\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m21\u001b[39m\n\u001b[0;32m 345\u001b[0m )\n\u001b[0;32m 346\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor4\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_60\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 347\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m10\u001b[39m\n\u001b[0;32m 348\u001b[0m )\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\groupby\\generic.py:230\u001b[0m, in \u001b[0;36mSeriesGroupBy.apply\u001b[1;34m(self, func, *args, **kwargs)\u001b[0m\n\u001b[0;32m 224\u001b[0m \u001b[38;5;129m@Appender\u001b[39m(\n\u001b[0;32m 225\u001b[0m _apply_docs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtemplate\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mformat(\n\u001b[0;32m 226\u001b[0m \u001b[38;5;28minput\u001b[39m\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mseries\u001b[39m\u001b[38;5;124m\"\u001b[39m, examples\u001b[38;5;241m=\u001b[39m_apply_docs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mseries_examples\u001b[39m\u001b[38;5;124m\"\u001b[39m]\n\u001b[0;32m 227\u001b[0m )\n\u001b[0;32m 228\u001b[0m )\n\u001b[0;32m 229\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mapply\u001b[39m(\u001b[38;5;28mself\u001b[39m, func, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Series:\n\u001b[1;32m--> 230\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39mapply(func, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\groupby\\groupby.py:1824\u001b[0m, in \u001b[0;36mGroupBy.apply\u001b[1;34m(self, func, include_groups, *args, **kwargs)\u001b[0m\n\u001b[0;32m 1822\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m option_context(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mmode.chained_assignment\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m):\n\u001b[0;32m 1823\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m-> 1824\u001b[0m result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_python_apply_general(f, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_selected_obj)\n\u001b[0;32m 1825\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (\n\u001b[0;32m 1826\u001b[0m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mobj, Series)\n\u001b[0;32m 1827\u001b[0m \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_selection \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[0;32m 1828\u001b[0m \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_selected_obj\u001b[38;5;241m.\u001b[39mshape \u001b[38;5;241m!=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_obj_with_exclusions\u001b[38;5;241m.\u001b[39mshape\n\u001b[0;32m 1829\u001b[0m ):\n\u001b[0;32m 1830\u001b[0m warnings\u001b[38;5;241m.\u001b[39mwarn(\n\u001b[0;32m 1831\u001b[0m message\u001b[38;5;241m=\u001b[39m_apply_groupings_depr\u001b[38;5;241m.\u001b[39mformat(\n\u001b[0;32m 1832\u001b[0m \u001b[38;5;28mtype\u001b[39m(\u001b[38;5;28mself\u001b[39m)\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mapply\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m (...)\u001b[0m\n\u001b[0;32m 1835\u001b[0m stacklevel\u001b[38;5;241m=\u001b[39mfind_stack_level(),\n\u001b[0;32m 1836\u001b[0m )\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\groupby\\groupby.py:1885\u001b[0m, in \u001b[0;36mGroupBy._python_apply_general\u001b[1;34m(self, f, data, not_indexed_same, is_transform, is_agg)\u001b[0m\n\u001b[0;32m 1850\u001b[0m \u001b[38;5;129m@final\u001b[39m\n\u001b[0;32m 1851\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_python_apply_general\u001b[39m(\n\u001b[0;32m 1852\u001b[0m \u001b[38;5;28mself\u001b[39m,\n\u001b[1;32m (...)\u001b[0m\n\u001b[0;32m 1857\u001b[0m is_agg: \u001b[38;5;28mbool\u001b[39m \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[0;32m 1858\u001b[0m ) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m NDFrameT:\n\u001b[0;32m 1859\u001b[0m \u001b[38;5;250m \u001b[39m\u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[0;32m 1860\u001b[0m \u001b[38;5;124;03m Apply function f in python space\u001b[39;00m\n\u001b[0;32m 1861\u001b[0m \n\u001b[1;32m (...)\u001b[0m\n\u001b[0;32m 1883\u001b[0m \u001b[38;5;124;03m data after applying f\u001b[39;00m\n\u001b[0;32m 1884\u001b[0m \u001b[38;5;124;03m \"\"\"\u001b[39;00m\n\u001b[1;32m-> 1885\u001b[0m values, mutated \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_grouper\u001b[38;5;241m.\u001b[39mapply_groupwise(f, data, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39maxis)\n\u001b[0;32m 1886\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m not_indexed_same \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m 1887\u001b[0m not_indexed_same \u001b[38;5;241m=\u001b[39m mutated\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\groupby\\ops.py:919\u001b[0m, in \u001b[0;36mBaseGrouper.apply_groupwise\u001b[1;34m(self, f, data, axis)\u001b[0m\n\u001b[0;32m 917\u001b[0m \u001b[38;5;66;03m# group might be modified\u001b[39;00m\n\u001b[0;32m 918\u001b[0m group_axes \u001b[38;5;241m=\u001b[39m group\u001b[38;5;241m.\u001b[39maxes\n\u001b[1;32m--> 919\u001b[0m res \u001b[38;5;241m=\u001b[39m f(group)\n\u001b[0;32m 920\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m mutated \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m _is_indexed_like(res, group_axes, axis):\n\u001b[0;32m 921\u001b[0m mutated \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n", - "File \u001b[1;32me:\\PyProject\\NewStock\\main\\train\\../..\\main\\factor\\factor.py:341\u001b[0m, in \u001b[0;36mget_rolling_factor..\u001b[1;34m(x)\u001b[0m\n\u001b[0;32m 336\u001b[0m \u001b[38;5;66;03m# 计算 act_factor1, act_factor2, act_factor3, act_factor4\u001b[39;00m\n\u001b[0;32m 337\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor1\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_5\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 338\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m50\u001b[39m\n\u001b[0;32m 339\u001b[0m )\n\u001b[0;32m 340\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor2\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_13\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[1;32m--> 341\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m40\u001b[39m\n\u001b[0;32m 342\u001b[0m )\n\u001b[0;32m 343\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor3\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_20\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 344\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m21\u001b[39m\n\u001b[0;32m 345\u001b[0m )\n\u001b[0;32m 346\u001b[0m df[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mact_factor4\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m grouped[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m_ema_60\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mapply(\n\u001b[0;32m 347\u001b[0m \u001b[38;5;28;01mlambda\u001b[39;00m x: np\u001b[38;5;241m.\u001b[39marctan((x \u001b[38;5;241m/\u001b[39m x\u001b[38;5;241m.\u001b[39mshift(\u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m-\u001b[39m \u001b[38;5;241m1\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m100\u001b[39m) \u001b[38;5;241m*\u001b[39m \u001b[38;5;241m57.3\u001b[39m \u001b[38;5;241m/\u001b[39m \u001b[38;5;241m10\u001b[39m\n\u001b[0;32m 348\u001b[0m )\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\ops\\common.py:76\u001b[0m, in \u001b[0;36m_unpack_zerodim_and_defer..new_method\u001b[1;34m(self, other)\u001b[0m\n\u001b[0;32m 72\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mNotImplemented\u001b[39m\n\u001b[0;32m 74\u001b[0m other \u001b[38;5;241m=\u001b[39m item_from_zerodim(other)\n\u001b[1;32m---> 76\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m method(\u001b[38;5;28mself\u001b[39m, other)\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\arraylike.py:194\u001b[0m, in \u001b[0;36mOpsMixin.__sub__\u001b[1;34m(self, other)\u001b[0m\n\u001b[0;32m 192\u001b[0m \u001b[38;5;129m@unpack_zerodim_and_defer\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m__sub__\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m 193\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__sub__\u001b[39m(\u001b[38;5;28mself\u001b[39m, other):\n\u001b[1;32m--> 194\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_arith_method(other, operator\u001b[38;5;241m.\u001b[39msub)\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\series.py:6135\u001b[0m, in \u001b[0;36mSeries._arith_method\u001b[1;34m(self, other, op)\u001b[0m\n\u001b[0;32m 6133\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m_arith_method\u001b[39m(\u001b[38;5;28mself\u001b[39m, other, op):\n\u001b[0;32m 6134\u001b[0m \u001b[38;5;28mself\u001b[39m, other \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_align_for_op(other)\n\u001b[1;32m-> 6135\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m base\u001b[38;5;241m.\u001b[39mIndexOpsMixin\u001b[38;5;241m.\u001b[39m_arith_method(\u001b[38;5;28mself\u001b[39m, other, op)\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\base.py:1382\u001b[0m, in \u001b[0;36mIndexOpsMixin._arith_method\u001b[1;34m(self, other, op)\u001b[0m\n\u001b[0;32m 1379\u001b[0m rvalues \u001b[38;5;241m=\u001b[39m np\u001b[38;5;241m.\u001b[39marange(rvalues\u001b[38;5;241m.\u001b[39mstart, rvalues\u001b[38;5;241m.\u001b[39mstop, rvalues\u001b[38;5;241m.\u001b[39mstep)\n\u001b[0;32m 1381\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m np\u001b[38;5;241m.\u001b[39merrstate(\u001b[38;5;28mall\u001b[39m\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mignore\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m-> 1382\u001b[0m result \u001b[38;5;241m=\u001b[39m ops\u001b[38;5;241m.\u001b[39marithmetic_op(lvalues, rvalues, op)\n\u001b[0;32m 1384\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_construct_result(result, name\u001b[38;5;241m=\u001b[39mres_name)\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\ops\\array_ops.py:283\u001b[0m, in \u001b[0;36marithmetic_op\u001b[1;34m(left, right, op)\u001b[0m\n\u001b[0;32m 279\u001b[0m _bool_arith_check(op, left, right) \u001b[38;5;66;03m# type: ignore[arg-type]\u001b[39;00m\n\u001b[0;32m 281\u001b[0m \u001b[38;5;66;03m# error: Argument 1 to \"_na_arithmetic_op\" has incompatible type\u001b[39;00m\n\u001b[0;32m 282\u001b[0m \u001b[38;5;66;03m# \"Union[ExtensionArray, ndarray[Any, Any]]\"; expected \"ndarray[Any, Any]\"\u001b[39;00m\n\u001b[1;32m--> 283\u001b[0m res_values \u001b[38;5;241m=\u001b[39m _na_arithmetic_op(left, right, op) \u001b[38;5;66;03m# type: ignore[arg-type]\u001b[39;00m\n\u001b[0;32m 285\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m res_values\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\ops\\array_ops.py:218\u001b[0m, in \u001b[0;36m_na_arithmetic_op\u001b[1;34m(left, right, op, is_cmp)\u001b[0m\n\u001b[0;32m 215\u001b[0m func \u001b[38;5;241m=\u001b[39m partial(expressions\u001b[38;5;241m.\u001b[39mevaluate, op)\n\u001b[0;32m 217\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m--> 218\u001b[0m result \u001b[38;5;241m=\u001b[39m func(left, right)\n\u001b[0;32m 219\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m:\n\u001b[0;32m 220\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_cmp \u001b[38;5;129;01mand\u001b[39;00m (\n\u001b[0;32m 221\u001b[0m left\u001b[38;5;241m.\u001b[39mdtype \u001b[38;5;241m==\u001b[39m \u001b[38;5;28mobject\u001b[39m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mgetattr\u001b[39m(right, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mdtype\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m==\u001b[39m \u001b[38;5;28mobject\u001b[39m\n\u001b[0;32m 222\u001b[0m ):\n\u001b[1;32m (...)\u001b[0m\n\u001b[0;32m 225\u001b[0m \u001b[38;5;66;03m# Don't do this for comparisons, as that will handle complex numbers\u001b[39;00m\n\u001b[0;32m 226\u001b[0m \u001b[38;5;66;03m# incorrectly, see GH#32047\u001b[39;00m\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\computation\\expressions.py:242\u001b[0m, in \u001b[0;36mevaluate\u001b[1;34m(op, a, b, use_numexpr)\u001b[0m\n\u001b[0;32m 239\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m op_str \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m 240\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m use_numexpr:\n\u001b[0;32m 241\u001b[0m \u001b[38;5;66;03m# error: \"None\" not callable\u001b[39;00m\n\u001b[1;32m--> 242\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m _evaluate(op, op_str, a, b) \u001b[38;5;66;03m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m 243\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m _evaluate_standard(op, op_str, a, b)\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\computation\\expressions.py:131\u001b[0m, in \u001b[0;36m_evaluate_numexpr\u001b[1;34m(op, op_str, a, b)\u001b[0m\n\u001b[0;32m 128\u001b[0m _store_test_result(result \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m)\n\u001b[0;32m 130\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m result \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m--> 131\u001b[0m result \u001b[38;5;241m=\u001b[39m _evaluate_standard(op, op_str, a, b)\n\u001b[0;32m 133\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m result\n", - "File \u001b[1;32me:\\Python\\anaconda\\envs\\new_trader\\Lib\\site-packages\\pandas\\core\\computation\\expressions.py:73\u001b[0m, in \u001b[0;36m_evaluate_standard\u001b[1;34m(op, op_str, a, b)\u001b[0m\n\u001b[0;32m 71\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m _TEST_MODE:\n\u001b[0;32m 72\u001b[0m _store_test_result(\u001b[38;5;28;01mFalse\u001b[39;00m)\n\u001b[1;32m---> 73\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m op(a, b)\n", - "\u001b[1;31mKeyboardInterrupt\u001b[0m: " + " dtype='object')\n", + "Calculating lg_flow_mom_corr_20_60...\n", + "Finished lg_flow_mom_corr_20_60.\n", + "Calculating lg_flow_accel...\n", + "Finished lg_flow_accel.\n", + "Calculating profit_pressure...\n", + "Finished profit_pressure.\n", + "Calculating underwater_resistance...\n", + "Finished underwater_resistance.\n", + "Calculating cost_conc_std_20...\n", + "Finished cost_conc_std_20.\n", + "Calculating profit_decay_20...\n", + "Finished profit_decay_20.\n", + "Calculating vol_amp_loss_20...\n", + "Finished vol_amp_loss_20.\n", + "Calculating vol_drop_profit_cnt_5...\n", + "Finished vol_drop_profit_cnt_5.\n", + "Calculating lg_flow_vol_interact_20...\n", + "Finished lg_flow_vol_interact_20.\n", + "Calculating cost_break_confirm_cnt_5...\n", + "Finished cost_break_confirm_cnt_5.\n", + "Calculating atr_norm_channel_pos_14...\n", + "Finished atr_norm_channel_pos_14.\n", + "Calculating turnover_diff_skew_20...\n", + "Finished turnover_diff_skew_20.\n", + "Calculating lg_sm_flow_diverge_20...\n", + "Finished lg_sm_flow_diverge_20.\n", + "Calculating pullback_strong_20_20...\n", + "Finished pullback_strong_20_20.\n", + "Calculating vol_wgt_hist_pos_20...\n", + "Finished vol_wgt_hist_pos_20.\n", + "Calculating vol_adj_roc_20...\n", + "Finished vol_adj_roc_20.\n", + "Calculating cs_rank_net_lg_flow_val...\n", + "Finished cs_rank_net_lg_flow_val.\n", + "Calculating cs_rank_flow_divergence...\n", + "Finished cs_rank_flow_divergence.\n", + "Calculating cs_rank_ind_adj_lg_flow...\n", + "Finished cs_rank_ind_adj_lg_flow.\n", + "Calculating cs_rank_elg_buy_ratio...\n", + "Finished cs_rank_elg_buy_ratio.\n", + "Calculating cs_rank_rel_profit_margin...\n", + "Finished cs_rank_rel_profit_margin.\n", + "Calculating cs_rank_cost_breadth...\n", + "Finished cs_rank_cost_breadth.\n", + "Calculating cs_rank_dist_to_upper_cost...\n", + "Finished cs_rank_dist_to_upper_cost.\n", + "Calculating cs_rank_winner_rate...\n", + "Finished cs_rank_winner_rate.\n", + "Calculating cs_rank_intraday_range...\n", + "Finished cs_rank_intraday_range.\n", + "Calculating cs_rank_close_pos_in_range...\n", + "Finished cs_rank_close_pos_in_range.\n", + "Calculating cs_rank_opening_gap...\n", + "Error calculating cs_rank_opening_gap: Missing 'pre_close' column. Assigning NaN.\n", + "Calculating cs_rank_pos_in_hist_range...\n", + "Finished cs_rank_pos_in_hist_range.\n", + "Calculating cs_rank_vol_x_profit_margin...\n", + "Finished cs_rank_vol_x_profit_margin.\n", + "Calculating cs_rank_lg_flow_price_concordance...\n", + "Finished cs_rank_lg_flow_price_concordance.\n", + "Calculating cs_rank_turnover_per_winner...\n", + "Finished cs_rank_turnover_per_winner.\n", + "Calculating cs_rank_ind_cap_neutral_pe (Placeholder - requires statsmodels)...\n", + "Finished cs_rank_ind_cap_neutral_pe (Placeholder).\n", + "Calculating cs_rank_volume_ratio...\n", + "Finished cs_rank_volume_ratio.\n", + "Calculating cs_rank_elg_buy_sell_sm_ratio...\n", + "Finished cs_rank_elg_buy_sell_sm_ratio.\n", + "Calculating cs_rank_cost_dist_vol_ratio...\n", + "Finished cs_rank_cost_dist_vol_ratio.\n", + "Calculating cs_rank_size...\n", + "Finished cs_rank_size.\n", + "\n", + "RangeIndex: 4539678 entries, 0 to 4539677\n", + "Columns: 178 entries, ts_code to cs_rank_size\n", + "dtypes: bool(10), datetime64[ns](1), float64(162), int32(3), object(2)\n", + "memory usage: 5.7+ GB\n", + "None\n", + "['ts_code', 'trade_date', 'open', 'close', 'high', 'low', 'vol', 'pct_chg', 'turnover_rate', 'pe_ttm', 'circ_mv', 'total_mv', 'volume_ratio', 'is_st', 'up_limit', 'down_limit', 'buy_sm_vol', 'sell_sm_vol', 'buy_lg_vol', 'sell_lg_vol', 'buy_elg_vol', 'sell_elg_vol', 'net_mf_vol', 'his_low', 'his_high', 'cost_5pct', 'cost_15pct', 'cost_50pct', 'cost_85pct', 'cost_95pct', 'weight_avg', 'winner_rate', 'cat_l2_code', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'flow_divergence_diff', 'flow_divergence_ratio', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'price_cost_divergence', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_flow_divergence', 'cs_rank_ind_adj_lg_flow', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_opening_gap', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_ind_cap_neutral_pe', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size']\n" ] } ], @@ -565,6 +624,15 @@ "# df = cat_reason(df, top_list_df)\n", "# df = cat_is_on_top_list(df, top_list_df)\n", "\n", + "df = ts_turnover_rate_acceleration_5_20(df)\n", + "df = ts_vol_sustain_10_30(df)\n", + "df = cs_turnover_rate_relative_strength_20(df)\n", + "df = cs_amount_outlier_10(df)\n", + "df = ts_ff_to_total_turnover_ratio(df)\n", + "df = ts_price_volume_trend_coherence_5_20(df)\n", + "df = ts_turnover_rate_trend_strength_5(df)\n", + "df = ts_ff_turnover_rate_surge_10(df)\n", + "\n", "df = add_financial_factor(df, fina_indicator_df, factor_value_col='undist_profit_ps')\n", "df = add_financial_factor(df, fina_indicator_df, factor_value_col='ocfps')\n", "calculate_arbr(df, N=26)\n", @@ -629,7 +697,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 10, "id": "b87b938028afa206", "metadata": { "ExecuteTime": { @@ -667,7 +735,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, "id": "f4f16d63ad18d1bc", "metadata": { "ExecuteTime": { @@ -913,7 +981,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "id": "40e6b68a91b30c79", "metadata": { "ExecuteTime": { @@ -1240,7 +1308,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 13, "id": "47c12bb34062ae7a", "metadata": { "ExecuteTime": { @@ -1262,7 +1330,7 @@ "# df['future_return'] = (df.groupby('ts_code')['close'].shift(-days) - df.groupby('ts_code')['open'].shift(-1)) / \\\n", "# df.groupby('ts_code')['open'].shift(-1)\n", "\n", - "df['cat_up_limit'] = df['pct_chg'] > 5\n", + "# df['cat_up_limit'] = df['pct_chg'] > 5\n", "# df['label'] = (df.groupby('ts_code')['cat_up_limit']\n", "# .rolling(window=5, min_periods=1).sum()\n", "# .groupby('ts_code') # 再次按 ts_code 分组\n", @@ -1271,7 +1339,7 @@ "# .astype(int)\n", "# .reset_index(level=0, drop=True))\n", "df['label'] = df.groupby('trade_date', group_keys=False)['future_return'].transform(\n", - " lambda x: pd.qcut(x, q=20, labels=False, duplicates='drop')\n", + " lambda x: pd.qcut(x, q=50, labels=False, duplicates='drop')\n", ")\n", "filter_index = df['future_return'].between(df['future_return'].quantile(0.01), df['future_return'].quantile(0.99))\n", "\n", @@ -1282,103 +1350,7 @@ }, { "cell_type": "code", - "execution_count": null, - "id": "8f4dc587", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - " ts_code trade_date open close high low vol pct_chg \\\n", - "976171 002193.SZ 2019-01-02 18.02 17.82 18.21 17.73 22513.62 -1.00 \n", - "976172 002193.SZ 2019-01-03 17.82 17.78 17.94 17.65 19523.21 -0.22 \n", - "976173 002193.SZ 2019-01-04 17.47 18.04 18.10 17.34 28094.69 1.46 \n", - "976174 002193.SZ 2019-01-07 18.17 18.31 18.39 18.00 23866.88 1.50 \n", - "976175 002193.SZ 2019-01-08 18.25 18.52 18.66 18.13 22853.46 1.15 \n", - "... ... ... ... ... ... ... ... ... \n", - "977704 002193.SZ 2025-05-19 11.74 11.95 11.97 11.61 94249.32 2.49 \n", - "977705 002193.SZ 2025-05-20 11.99 12.20 12.22 11.84 103040.57 2.09 \n", - "977706 002193.SZ 2025-05-21 12.20 11.97 12.28 11.82 83112.00 -1.89 \n", - "977707 002193.SZ 2025-05-22 11.95 11.80 12.17 11.72 88811.00 -1.42 \n", - "977708 002193.SZ 2025-05-23 11.74 11.51 11.90 11.51 93013.00 -2.46 \n", - "\n", - " turnover_rate pe_ttm ... cs_rank_lg_flow_price_concordance \\\n", - "976171 0.9771 23.0129 ... 0.621118 \n", - "976172 0.8473 22.9598 ... 0.230095 \n", - "976173 1.2193 23.3049 ... 0.507119 \n", - "976174 1.0358 23.6499 ... 0.315078 \n", - "976175 0.9918 23.9154 ... 0.248631 \n", - "... ... ... ... ... \n", - "977704 3.6012 NaN ... 0.224402 \n", - "977705 3.9371 NaN ... 0.124834 \n", - "977706 3.1757 NaN ... 0.803720 \n", - "977707 3.3934 NaN ... 0.667220 \n", - "977708 3.5540 NaN ... 0.308867 \n", - "\n", - " cs_rank_turnover_per_winner cs_rank_ind_cap_neutral_pe \\\n", - "976171 0.990869 NaN \n", - "976172 0.934283 NaN \n", - "976173 0.925182 NaN \n", - "976174 0.843796 NaN \n", - "976175 0.834672 NaN \n", - "... ... ... \n", - "977704 0.624335 NaN \n", - "977705 0.641102 NaN \n", - "977706 0.624709 NaN \n", - "977707 0.563268 NaN \n", - "977708 0.661242 NaN \n", - "\n", - " cs_rank_volume_ratio cs_rank_elg_buy_sell_sm_ratio \\\n", - "976171 0.710344 0.341855 \n", - "976172 0.444444 0.318912 \n", - "976173 0.489226 0.260036 \n", - "976174 0.250000 0.251095 \n", - "976175 0.510588 0.286679 \n", - "... ... ... \n", - "977704 0.234043 0.397274 \n", - "977705 0.344124 0.116534 \n", - "977706 0.221189 0.126370 \n", - "977707 0.412155 0.130521 \n", - "977708 0.534540 0.134175 \n", - "\n", - " cs_rank_cost_dist_vol_ratio cs_rank_size future_return \\\n", - "976171 0.261603 0.262235 0.034231 \n", - "976172 0.185342 0.264695 0.029809 \n", - "976173 0.211959 0.259489 0.025499 \n", - "976174 0.145266 0.255474 0.014746 \n", - "976175 0.202299 0.259854 0.003240 \n", - "... ... ... ... \n", - "977704 0.299535 0.031250 NaN \n", - "977705 0.339641 0.032205 NaN \n", - "977706 0.291597 0.030555 NaN \n", - "977707 0.372634 0.030555 NaN \n", - "977708 0.430090 0.027898 NaN \n", - "\n", - " cat_up_limit label \n", - "976171 False 8.0 \n", - "976172 False 7.0 \n", - "976173 False 10.0 \n", - "976174 False 13.0 \n", - "976175 False 6.0 \n", - "... ... ... \n", - "977704 False NaN \n", - "977705 False NaN \n", - "977706 False NaN \n", - "977707 False NaN \n", - "977708 False NaN \n", - "\n", - "[1538 rows x 181 columns]\n" - ] - } - ], - "source": [ - "print(df[df['ts_code'] == '002193.SZ'])" - ] - }, - { - "cell_type": "code", - "execution_count": null, + "execution_count": 14, "id": "29221dde", "metadata": {}, "outputs": [ @@ -1386,7 +1358,7 @@ "name": "stdout", "output_type": "stream", "text": [ - "192\n" + "191\n" ] } ], @@ -1425,7 +1397,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 15, "id": "03ee5daf", "metadata": {}, "outputs": [], @@ -1438,7 +1410,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 16, "id": "b76ea08a", "metadata": {}, "outputs": [ @@ -1450,7 +1422,7 @@ "0 000001.SZ 2019-01-02 16.574219\n", "1 000001.SZ 2019-01-03 16.583965\n", "2 000001.SZ 2019-01-04 16.633371\n", - "['vol', 'pct_chg', 'turnover_rate', 'volume_ratio', 'winner_rate', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size', 'cat_up_limit', 'industry_obv', 'industry_return_5', 'industry_return_20', 'industry__ema_5', 'industry__ema_13', 'industry__ema_20', 'industry__ema_60', 'industry_act_factor1', 'industry_act_factor2', 'industry_act_factor3', 'industry_act_factor4', 'industry_act_factor5', 'industry_act_factor6', 'industry_rank_act_factor1', 'industry_rank_act_factor2', 'industry_rank_act_factor3', 'industry_return_5_percentile', 'industry_return_20_percentile', '000852.SH_MACD', '000905.SH_MACD', '399006.SZ_MACD', '000852.SH_MACD_hist', '000905.SH_MACD_hist', '399006.SZ_MACD_hist', '000852.SH_RSI', '000905.SH_RSI', '399006.SZ_RSI', '000852.SH_Signal_line', '000905.SH_Signal_line', '399006.SZ_Signal_line', '000852.SH_amount_change_rate', '000905.SH_amount_change_rate', '399006.SZ_amount_change_rate', '000852.SH_amount_mean', '000905.SH_amount_mean', '399006.SZ_amount_mean', '000852.SH_daily_return', '000905.SH_daily_return', '399006.SZ_daily_return', '000852.SH_up_ratio_20d', '000905.SH_up_ratio_20d', '399006.SZ_up_ratio_20d', '000852.SH_volatility', '000905.SH_volatility', '399006.SZ_volatility', '000852.SH_volume_change_rate', '000905.SH_volume_change_rate', '399006.SZ_volume_change_rate']\n", + "['vol', 'pct_chg', 'turnover_rate', 'volume_ratio', 'winner_rate', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size', 'industry_obv', 'industry_return_5', 'industry_return_20', 'industry__ema_5', 'industry__ema_13', 'industry__ema_20', 'industry__ema_60', 'industry_act_factor1', 'industry_act_factor2', 'industry_act_factor3', 'industry_act_factor4', 'industry_act_factor5', 'industry_act_factor6', 'industry_rank_act_factor1', 'industry_rank_act_factor2', 'industry_rank_act_factor3', 'industry_return_5_percentile', 'industry_return_20_percentile', '000852.SH_MACD', '000905.SH_MACD', '399006.SZ_MACD', '000852.SH_MACD_hist', '000905.SH_MACD_hist', '399006.SZ_MACD_hist', '000852.SH_RSI', '000905.SH_RSI', '399006.SZ_RSI', '000852.SH_Signal_line', '000905.SH_Signal_line', '399006.SZ_Signal_line', '000852.SH_amount_change_rate', '000905.SH_amount_change_rate', '399006.SZ_amount_change_rate', '000852.SH_amount_mean', '000905.SH_amount_mean', '399006.SZ_amount_mean', '000852.SH_daily_return', '000905.SH_daily_return', '399006.SZ_daily_return', '000852.SH_up_ratio_20d', '000905.SH_up_ratio_20d', '399006.SZ_up_ratio_20d', '000852.SH_volatility', '000905.SH_volatility', '399006.SZ_volatility', '000852.SH_volume_change_rate', '000905.SH_volume_change_rate', '399006.SZ_volume_change_rate']\n", "去除极值\n", "开始截面 MAD 去极值处理 (k=3.0)...\n" ] @@ -1474,7 +1446,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "MAD Filtering: 100%|██████████| 132/132 [00:23<00:00, 5.57it/s]\n" + "MAD Filtering: 100%|██████████| 132/132 [00:24<00:00, 5.45it/s]\n" ] }, { @@ -1512,7 +1484,7 @@ "output_type": "stream", "text": [ "截面 MAD 去极值处理完成。\n", - "feature_columns: ['vol', 'pct_chg', 'turnover_rate', 'volume_ratio', 'winner_rate', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size', 'cat_up_limit', 'industry_obv', 'industry_return_5', 'industry_return_20', 'industry__ema_5', 'industry__ema_13', 'industry__ema_20', 'industry__ema_60', 'industry_act_factor1', 'industry_act_factor2', 'industry_act_factor3', 'industry_act_factor4', 'industry_act_factor5', 'industry_act_factor6', 'industry_rank_act_factor1', 'industry_rank_act_factor2', 'industry_rank_act_factor3', 'industry_return_5_percentile', 'industry_return_20_percentile', '000852.SH_MACD', '000905.SH_MACD', '399006.SZ_MACD', '000852.SH_MACD_hist', '000905.SH_MACD_hist', '399006.SZ_MACD_hist', '000852.SH_RSI', '000905.SH_RSI', '399006.SZ_RSI', '000852.SH_Signal_line', '000905.SH_Signal_line', '399006.SZ_Signal_line', '000852.SH_amount_change_rate', '000905.SH_amount_change_rate', '399006.SZ_amount_change_rate', '000852.SH_amount_mean', '000905.SH_amount_mean', '399006.SZ_amount_mean', '000852.SH_daily_return', '000905.SH_daily_return', '399006.SZ_daily_return', '000852.SH_up_ratio_20d', '000905.SH_up_ratio_20d', '399006.SZ_up_ratio_20d', '000852.SH_volatility', '000905.SH_volatility', '399006.SZ_volatility', '000852.SH_volume_change_rate', '000905.SH_volume_change_rate', '399006.SZ_volume_change_rate']\n", + "feature_columns: ['vol', 'pct_chg', 'turnover_rate', 'volume_ratio', 'winner_rate', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size', 'industry_obv', 'industry_return_5', 'industry_return_20', 'industry__ema_5', 'industry__ema_13', 'industry__ema_20', 'industry__ema_60', 'industry_act_factor1', 'industry_act_factor2', 'industry_act_factor3', 'industry_act_factor4', 'industry_act_factor5', 'industry_act_factor6', 'industry_rank_act_factor1', 'industry_rank_act_factor2', 'industry_rank_act_factor3', 'industry_return_5_percentile', 'industry_return_20_percentile', '000852.SH_MACD', '000905.SH_MACD', '399006.SZ_MACD', '000852.SH_MACD_hist', '000905.SH_MACD_hist', '399006.SZ_MACD_hist', '000852.SH_RSI', '000905.SH_RSI', '399006.SZ_RSI', '000852.SH_Signal_line', '000905.SH_Signal_line', '399006.SZ_Signal_line', '000852.SH_amount_change_rate', '000905.SH_amount_change_rate', '399006.SZ_amount_change_rate', '000852.SH_amount_mean', '000905.SH_amount_mean', '399006.SZ_amount_mean', '000852.SH_daily_return', '000905.SH_daily_return', '399006.SZ_daily_return', '000852.SH_up_ratio_20d', '000905.SH_up_ratio_20d', '399006.SZ_up_ratio_20d', '000852.SH_volatility', '000905.SH_volatility', '399006.SZ_volatility', '000852.SH_volume_change_rate', '000905.SH_volume_change_rate', '399006.SZ_volume_change_rate']\n", "df最小日期: 2019-01-02\n", "df最大日期: 2025-05-23\n", "2057539\n", @@ -1656,57 +1628,7 @@ }, { "cell_type": "code", - "execution_count": null, - "id": "2e4b027e", - "metadata": {}, - "outputs": [], - "source": [ - "class RmseObjective(object):\n", - "\n", - " def is_max_optimal(self):\n", - " return False\n", - "\n", - " def get_final_error(self, error, weight):\n", - " return np.sqrt(error / (weight + 1e-38))\n", - "\n", - "\n", - " def evaluate(self, approxes, target, weight):\n", - " assert len(approxes) == 1\n", - " assert len(target) == len(approxes[0])\n", - "\n", - " approx = approxes[0]\n", - "\n", - " error_sum = 0.0\n", - " weight_sum = 0.0\n", - "\n", - " for i in range(len(approx)):\n", - " w = 1.0 if weight is None else weight[i]\n", - " weight_sum += w\n", - " error_sum += w * ((approx[i] - target[i])**2)\n", - "\n", - " return error_sum, weight_sum\n", - "\n", - " def calc_ders_range(self, approxes, targets, weights):\n", - " assert len(approxes) == len(targets)\n", - " if weights is not None:\n", - " assert len(weights) == len(approxes)\n", - "\n", - " result = []\n", - " for index in range(len(targets)):\n", - " der1 = targets[index] - approxes[index]\n", - " der2 = -1\n", - "\n", - " if weights is not None:\n", - " der1 *= weights[index]\n", - " der2 *= weights[index]\n", - "\n", - " result.append((der1, der2))\n", - " return result\n" - ] - }, - { - "cell_type": "code", - "execution_count": null, + "execution_count": 100, "id": "3ff2d1c5", "metadata": {}, "outputs": [], @@ -1817,24 +1739,24 @@ " params = {\n", " 'label_gain': [gain * gain for gain in label_gain],\n", " 'objective': 'lambdarank',\n", - " 'metric': 'rank_xendcg',\n", + " 'metric': 'lambdarank',\n", " 'learning_rate': 0.01,\n", " 'num_leaves': 1024,\n", - " # 'min_data_in_leaf': 128,\n", - " 'max_depth': 10,\n", + " 'min_data_in_leaf': 256,\n", + " # 'max_depth': 64,\n", " 'max_bin': 1024,\n", " 'feature_fraction': 0.7,\n", " 'bagging_fraction': 0.7,\n", " 'bagging_freq': 5,\n", - " 'lambda_l1': 1,\n", - " 'lambda_l2': 1,\n", + " 'lambda_l1': 10,\n", + " # 'lambda_l2': 1,\n", " 'boosting': 'gbdt',\n", " 'verbosity': -1,\n", " 'extra_trees': True,\n", " # 'max_position': 5,\n", - " # 'ndcg_at': 1,\n", + " 'ndcg_at': 5,\n", " 'quant_train_renew_leaf': True,\n", - " # 'lambdarank_truncation_level': 1,\n", + " 'lambdarank_truncation_level': 10,\n", " # 'lambdarank_position_bias_regularization': 1,\n", " 'seed': 7\n", " }\n", @@ -1856,44 +1778,44 @@ " evals = {}\n", " callbacks = [lgb.log_evaluation(period=1000),\n", " lgb.callback.record_evaluation(evals),\n", - " lgb.early_stopping(100, first_metric_only=True)\n", + " lgb.early_stopping(500, first_metric_only=False)\n", " ]\n", - " # # 训练模型\n", - " # model = lgb.train(\n", - " # params, train_dataset, num_boost_round=1000,\n", - " # valid_sets=[train_dataset, val_dataset], valid_names=['train', 'valid'],\n", - " # callbacks=callbacks\n", - " # )\n", + " # 训练模型\n", + " model = lgb.train(\n", + " params, train_dataset, num_boost_round=1000,\n", + " valid_sets=[train_dataset, val_dataset], valid_names=['train', 'valid'],\n", + " callbacks=callbacks\n", + " )\n", "\n", - " # # 打印特征重要性（如果需要）\n", - " # if True:\n", - " # lgb.plot_metric(evals)\n", - " # lgb.plot_importance(model, importance_type='split', max_num_features=20)\n", - " # plt.show()\n", + " # 打印特征重要性（如果需要）\n", + " if True:\n", + " lgb.plot_metric(evals)\n", + " lgb.plot_importance(model, importance_type='split', max_num_features=20)\n", + " plt.show()\n", "\n", - " from flaml import AutoML\n", - " from sklearn.datasets import fetch_california_housing\n", + " # from flaml import AutoML\n", + " # from sklearn.datasets import fetch_california_housing\n", "\n", - " # Initialize an AutoML instance\n", - " model = AutoML()\n", - " # Specify automl goal and constraint\n", - " automl_settings = {\n", - " \"time_budget\": 600, # in seconds\n", - " \"metric\": \"ndcg@1\",\n", - " \"task\": \"rank\",\n", - " \"estimator_list\": [\n", - " \"catboost\",\n", - " \"lgbm\",\n", - " \"xgboost\"\n", - " ], \n", - " \"ensemble\": {\n", - " \"final_estimator\": LGBMRanker(),\n", - " \"passthrough\": False,\n", - " },\n", - " }\n", - " model.fit(X_train=X_train, y_train=y_train, groups=train_groups,\n", - " X_val=X_val, y_val=y_val,groups_val=val_groups,\n", - " mlflow_logging=False, **automl_settings)\n", + " # # Initialize an AutoML instance\n", + " # model = AutoML()\n", + " # # Specify automl goal and constraint\n", + " # automl_settings = {\n", + " # \"time_budget\": 600, # in seconds\n", + " # \"metric\": \"ndcg@1\",\n", + " # \"task\": \"rank\",\n", + " # \"estimator_list\": [\n", + " # \"catboost\",\n", + " # \"lgbm\",\n", + " # \"xgboost\"\n", + " # ], \n", + " # \"ensemble\": {\n", + " # \"final_estimator\": LGBMRanker(),\n", + " # \"passthrough\": False,\n", + " # },\n", + " # }\n", + " # model.fit(X_train=X_train, y_train=y_train, groups=train_groups,\n", + " # X_val=X_val, y_val=y_val,groups_val=val_groups,\n", + " # mlflow_logging=False, **automl_settings)\n", "\n", "\n", " return model, scaler, None # 返回训练好的模型、scaler 和 pca 对象" @@ -1901,7 +1823,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 101, "id": "c6eb5cd4-e714-420a-ac48-39af3e11ee81", "metadata": { "ExecuteTime": { @@ -1914,116 +1836,46 @@ "name": "stdout", "output_type": "stream", "text": [ - "train data size: 728000\n", - " ts_code trade_date log_circ_mv\n", - "0 600306.SH 2020-01-02 11.552040\n", - "1 603269.SH 2020-01-02 11.324801\n", - "2 002633.SZ 2020-01-02 11.759023\n", - "3 603991.SH 2020-01-02 11.181150\n", - "4 000691.SZ 2020-01-02 11.677910\n", - "... ... ... ...\n", - "727995 002235.SZ 2022-12-30 12.914708\n", - "727996 605598.SH 2022-12-30 11.783580\n", - "727997 002613.SZ 2022-12-30 12.489464\n", - "727998 600800.SH 2022-12-30 12.571911\n", - "727999 603068.SH 2022-12-30 12.967134\n", + "train data size: 1091832\n", + " ts_code trade_date log_circ_mv\n", + "0 600306.SH 2020-01-02 11.552040\n", + "1 603269.SH 2020-01-02 11.324801\n", + "2 002633.SZ 2020-01-02 11.759023\n", + "3 603991.SH 2020-01-02 11.181150\n", + "4 000691.SZ 2020-01-02 11.677910\n", + "... ... ... ...\n", + "1091827 603698.SH 2022-12-30 13.370853\n", + "1091828 600789.SH 2022-12-30 13.372063\n", + "1091829 605366.SH 2022-12-30 12.642936\n", + "1091830 603219.SH 2022-12-30 12.089671\n", + "1091831 000615.SZ 2022-12-30 13.375555\n", "\n", - "[728000 rows x 3 columns]\n", - "原始样本数: 728000, 去除标签为空后样本数: 728000\n", - "[flaml.automl.logger: 05-26 20:16:56] {1728} INFO - task = rank\n", - "[flaml.automl.logger: 05-26 20:16:56] {1736} INFO - Data split method: group\n", - "[flaml.automl.logger: 05-26 20:16:56] {1739} INFO - Evaluation method: holdout\n", - "[flaml.automl.logger: 05-26 20:16:56] {1838} INFO - Minimizing error metric: 1-ndcg@1\n", - "[flaml.automl.logger: 05-26 20:16:56] {1955} INFO - List of ML learners in AutoML Run: ['catboost', 'lgbm', 'xgboost']\n", - "[flaml.automl.logger: 05-26 20:16:56] {2258} INFO - iteration 0, current learner catboost\n", - "[flaml.automl.logger: 05-26 20:17:16] {2393} INFO - Estimated sufficient time budget=203147s. Estimated necessary time budget=203s.\n", - "[flaml.automl.logger: 05-26 20:17:16] {2442} INFO - at 44.5s,\testimator catboost's best error=0.3789,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:17:16] {2258} INFO - iteration 1, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:17:20] {2442} INFO - at 48.3s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:17:20] {2258} INFO - iteration 2, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:17:25] {2442} INFO - at 53.2s,\testimator xgboost's best error=0.5269,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:17:25] {2258} INFO - iteration 3, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:17:29] {2442} INFO - at 57.0s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:17:29] {2258} INFO - iteration 4, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:17:34] {2442} INFO - at 61.6s,\testimator xgboost's best error=0.5269,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:17:34] {2258} INFO - iteration 5, current learner catboost\n", - "[flaml.automl.logger: 05-26 20:19:53] {2442} INFO - at 200.7s,\testimator catboost's best error=0.3789,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:19:53] {2258} INFO - iteration 6, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:19:56] {2442} INFO - at 204.4s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:19:56] {2258} INFO - iteration 7, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:20:01] {2442} INFO - at 209.4s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:01] {2258} INFO - iteration 8, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:20:05] {2442} INFO - at 212.9s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:05] {2258} INFO - iteration 9, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:20:09] {2442} INFO - at 216.6s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:09] {2258} INFO - iteration 10, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:13] {2442} INFO - at 221.4s,\testimator xgboost's best error=0.5266,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:13] {2258} INFO - iteration 11, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:18] {2442} INFO - at 226.2s,\testimator xgboost's best error=0.4896,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:18] {2258} INFO - iteration 12, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:23] {2442} INFO - at 231.0s,\testimator xgboost's best error=0.4896,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:23] {2258} INFO - iteration 13, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:28] {2442} INFO - at 235.7s,\testimator xgboost's best error=0.4896,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:28] {2258} INFO - iteration 14, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:34] {2442} INFO - at 242.2s,\testimator xgboost's best error=0.4896,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:34] {2258} INFO - iteration 15, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:20:38] {2442} INFO - at 246.3s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:38] {2258} INFO - iteration 16, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:44] {2442} INFO - at 251.8s,\testimator xgboost's best error=0.4896,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:44] {2258} INFO - iteration 17, current learner catboost\n", - "[flaml.automl.logger: 05-26 20:20:52] {2442} INFO - at 260.4s,\testimator catboost's best error=0.3789,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:52] {2258} INFO - iteration 18, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:20:58] {2442} INFO - at 265.8s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:20:58] {2258} INFO - iteration 19, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:21:02] {2442} INFO - at 269.7s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:02] {2258} INFO - iteration 20, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:21:07] {2442} INFO - at 275.0s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:07] {2258} INFO - iteration 21, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:21:12] {2442} INFO - at 279.9s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:12] {2258} INFO - iteration 22, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:21:15] {2442} INFO - at 283.5s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:15] {2258} INFO - iteration 23, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:21:20] {2442} INFO - at 288.1s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:20] {2258} INFO - iteration 24, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:21:25] {2442} INFO - at 292.8s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:25] {2258} INFO - iteration 25, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:21:30] {2442} INFO - at 298.0s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:30] {2258} INFO - iteration 26, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:21:36] {2442} INFO - at 304.0s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:36] {2258} INFO - iteration 27, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:21:41] {2442} INFO - at 309.0s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:41] {2258} INFO - iteration 28, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:21:48] {2442} INFO - at 316.1s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:21:48] {2258} INFO - iteration 29, current learner catboost\n", - "[flaml.automl.logger: 05-26 20:24:08] {2442} INFO - at 456.2s,\testimator catboost's best error=0.3789,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:08] {2258} INFO - iteration 30, current learner catboost\n", - "[flaml.automl.logger: 05-26 20:24:17] {2442} INFO - at 465.0s,\testimator catboost's best error=0.3789,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:17] {2258} INFO - iteration 31, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:24:22] {2442} INFO - at 469.9s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:22] {2258} INFO - iteration 32, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:24:27] {2442} INFO - at 474.9s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:27] {2258} INFO - iteration 33, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:24:30] {2442} INFO - at 478.6s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:30] {2258} INFO - iteration 34, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:24:35] {2442} INFO - at 483.1s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:35] {2258} INFO - iteration 35, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:24:41] {2442} INFO - at 489.2s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:41] {2258} INFO - iteration 36, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:24:46] {2442} INFO - at 493.8s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:46] {2258} INFO - iteration 37, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:24:52] {2442} INFO - at 500.5s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:52] {2258} INFO - iteration 38, current learner lgbm\n", - "[flaml.automl.logger: 05-26 20:24:56] {2442} INFO - at 504.3s,\testimator lgbm's best error=0.4831,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:24:56] {2258} INFO - iteration 39, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:25:01] {2442} INFO - at 509.0s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:25:01] {2258} INFO - iteration 40, current learner xgboost\n", - "[flaml.automl.logger: 05-26 20:25:09] {2442} INFO - at 516.7s,\testimator xgboost's best error=0.4722,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:25:09] {2258} INFO - iteration 41, current learner catboost\n", - "[flaml.automl.logger: 05-26 20:26:37] {2442} INFO - at 605.2s,\testimator catboost's best error=0.3789,\tbest estimator catboost's best error=0.3789\n", - "[flaml.automl.logger: 05-26 20:26:37] {2550} INFO - selected model: \n", - "[flaml.automl.logger: 05-26 20:26:37] {1985} INFO - fit succeeded\n", - "[flaml.automl.logger: 05-26 20:26:37] {1986} INFO - Time taken to find the best model: 44.485583543777466\n" + "[1091832 rows x 3 columns]\n", + "原始样本数: 1091832, 去除标签为空后样本数: 1091832\n", + "Training until validation scores don't improve for 500 rounds\n", + "Early stopping, best iteration is:\n", + "[3]\ttrain's ndcg@5: 0.406681\tvalid's ndcg@5: 0.383736\n" ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -2037,7 +1889,7 @@ "# print(f'feature_contri: {feature_contri}')\n", "model, scaler, pca = train_model(train_data\n", " .dropna(subset=['label']).groupby('trade_date', group_keys=False)\n", - " .apply(lambda x: x.nsmallest(1000, 'total_mv'))\n", + " .apply(lambda x: x.nsmallest(1500, 'total_mv'))\n", " .merge(industry_df, on=['cat_l2_code', 'trade_date'], how='left')\n", " .merge(index_data, on='trade_date', how='left'), \n", " feature_columns, type=type, target_column='label')\n" @@ -2045,7 +1897,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 102, "id": "5d1522a7538db91b", "metadata": { "ExecuteTime": { @@ -2087,7 +1939,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 103, "id": "09b1799e", "metadata": {}, "outputs": [ @@ -2095,8 +1947,8 @@ "name": "stdout", "output_type": "stream", "text": [ - "192\n", - "['vol', 'pct_chg', 'turnover_rate', 'volume_ratio', 'winner_rate', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size', 'cat_up_limit', 'industry_obv', 'industry_return_5', 'industry_return_20', 'industry__ema_5', 'industry__ema_13', 'industry__ema_20', 'industry__ema_60', 'industry_act_factor1', 'industry_act_factor2', 'industry_act_factor3', 'industry_act_factor4', 'industry_act_factor5', 'industry_act_factor6', 'industry_rank_act_factor1', 'industry_rank_act_factor2', 'industry_rank_act_factor3', 'industry_return_5_percentile', 'industry_return_20_percentile', '000852.SH_MACD', '000905.SH_MACD', '399006.SZ_MACD', '000852.SH_MACD_hist', '000905.SH_MACD_hist', '399006.SZ_MACD_hist', '000852.SH_RSI', '000905.SH_RSI', '399006.SZ_RSI', '000852.SH_Signal_line', '000905.SH_Signal_line', '399006.SZ_Signal_line', '000852.SH_amount_change_rate', '000905.SH_amount_change_rate', '399006.SZ_amount_change_rate', '000852.SH_amount_mean', '000905.SH_amount_mean', '399006.SZ_amount_mean', '000852.SH_daily_return', '000905.SH_daily_return', '399006.SZ_daily_return', '000852.SH_up_ratio_20d', '000905.SH_up_ratio_20d', '399006.SZ_up_ratio_20d', '000852.SH_volatility', '000905.SH_volatility', '399006.SZ_volatility', '000852.SH_volume_change_rate', '000905.SH_volume_change_rate', '399006.SZ_volume_change_rate']\n" + "191\n", + "['vol', 'pct_chg', 'turnover_rate', 'volume_ratio', 'winner_rate', 'undist_profit_ps', 'ocfps', 'AR', 'BR', 'AR_BR', 'log_circ_mv', 'cashflow_to_ev_factor', 'book_to_price_ratio', 'turnover_rate_mean_5', 'variance_20', 'bbi_ratio_factor', 'daily_deviation', 'lg_elg_net_buy_vol', 'flow_lg_elg_intensity', 'sm_net_buy_vol', 'total_buy_vol', 'lg_elg_buy_prop', 'flow_struct_buy_change', 'lg_elg_net_buy_vol_change', 'flow_lg_elg_accel', 'chip_concentration_range', 'chip_skewness', 'floating_chip_proxy', 'cost_support_15pct_change', 'cat_winner_price_zone', 'flow_chip_consistency', 'profit_taking_vs_absorb', 'cat_is_positive', 'upside_vol', 'downside_vol', 'vol_ratio', 'return_skew', 'return_kurtosis', 'volume_change_rate', 'cat_volume_breakout', 'turnover_deviation', 'cat_turnover_spike', 'avg_volume_ratio', 'cat_volume_ratio_breakout', 'vol_spike', 'vol_std_5', 'atr_14', 'atr_6', 'obv', 'maobv_6', 'rsi_3', 'return_5', 'return_20', 'std_return_5', 'std_return_90', 'std_return_90_2', 'act_factor1', 'act_factor2', 'act_factor3', 'act_factor4', 'rank_act_factor1', 'rank_act_factor2', 'rank_act_factor3', 'cov', 'delta_cov', 'alpha_22_improved', 'alpha_003', 'alpha_007', 'alpha_013', 'vol_break', 'weight_roc5', 'smallcap_concentration', 'cost_stability', 'high_cost_break_days', 'liquidity_risk', 'turnover_std', 'mv_volatility', 'volume_growth', 'mv_growth', 'momentum_factor', 'resonance_factor', 'log_close', 'cat_vol_spike', 'up', 'down', 'obv_maobv_6', 'std_return_5_over_std_return_90', 'std_return_90_minus_std_return_90_2', 'cat_af2', 'cat_af3', 'cat_af4', 'act_factor5', 'act_factor6', 'active_buy_volume_large', 'active_buy_volume_big', 'active_buy_volume_small', 'buy_lg_vol_minus_sell_lg_vol', 'buy_elg_vol_minus_sell_elg_vol', 'ctrl_strength', 'low_cost_dev', 'asymmetry', 'lock_factor', 'cat_vol_break', 'cost_atr_adj', 'cat_golden_resonance', 'mv_turnover_ratio', 'mv_adjusted_volume', 'mv_weighted_turnover', 'nonlinear_mv_volume', 'mv_volume_ratio', 'mv_momentum', 'lg_flow_mom_corr_20_60', 'lg_flow_accel', 'profit_pressure', 'underwater_resistance', 'cost_conc_std_20', 'profit_decay_20', 'vol_amp_loss_20', 'vol_drop_profit_cnt_5', 'lg_flow_vol_interact_20', 'cost_break_confirm_cnt_5', 'atr_norm_channel_pos_14', 'turnover_diff_skew_20', 'lg_sm_flow_diverge_20', 'pullback_strong_20_20', 'vol_wgt_hist_pos_20', 'vol_adj_roc_20', 'cs_rank_net_lg_flow_val', 'cs_rank_elg_buy_ratio', 'cs_rank_rel_profit_margin', 'cs_rank_cost_breadth', 'cs_rank_dist_to_upper_cost', 'cs_rank_winner_rate', 'cs_rank_intraday_range', 'cs_rank_close_pos_in_range', 'cs_rank_pos_in_hist_range', 'cs_rank_vol_x_profit_margin', 'cs_rank_lg_flow_price_concordance', 'cs_rank_turnover_per_winner', 'cs_rank_volume_ratio', 'cs_rank_elg_buy_sell_sm_ratio', 'cs_rank_cost_dist_vol_ratio', 'cs_rank_size', 'industry_obv', 'industry_return_5', 'industry_return_20', 'industry__ema_5', 'industry__ema_13', 'industry__ema_20', 'industry__ema_60', 'industry_act_factor1', 'industry_act_factor2', 'industry_act_factor3', 'industry_act_factor4', 'industry_act_factor5', 'industry_act_factor6', 'industry_rank_act_factor1', 'industry_rank_act_factor2', 'industry_rank_act_factor3', 'industry_return_5_percentile', 'industry_return_20_percentile', '000852.SH_MACD', '000905.SH_MACD', '399006.SZ_MACD', '000852.SH_MACD_hist', '000905.SH_MACD_hist', '399006.SZ_MACD_hist', '000852.SH_RSI', '000905.SH_RSI', '399006.SZ_RSI', '000852.SH_Signal_line', '000905.SH_Signal_line', '399006.SZ_Signal_line', '000852.SH_amount_change_rate', '000905.SH_amount_change_rate', '399006.SZ_amount_change_rate', '000852.SH_amount_mean', '000905.SH_amount_mean', '399006.SZ_amount_mean', '000852.SH_daily_return', '000905.SH_daily_return', '399006.SZ_daily_return', '000852.SH_up_ratio_20d', '000905.SH_up_ratio_20d', '399006.SZ_up_ratio_20d', '000852.SH_volatility', '000905.SH_volatility', '399006.SZ_volatility', '000852.SH_volume_change_rate', '000905.SH_volume_change_rate', '399006.SZ_volume_change_rate']\n" ] } ], @@ -2107,7 +1959,7 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 104, "id": "bceabd1f", "metadata": {}, "outputs": [ @@ -2118,7 +1970,7 @@ "警告: DataFrame 中没有 'group_id' 列。假设整个 DataFrame 是一个需要排序的组。\n", "\n", "NDCG 结果\n", - "{'ndcg@1': 0.16666666666666666, 'ndcg@3': 0.12073315639204953, 'ndcg@5': 0.16262598210797247}\n" + "{'ndcg@1': 0.9795918367346939, 'ndcg@3': 0.7667326972309916, 'ndcg@5': 0.6789315367339909}\n" ] } ], diff --git a/main/train/catboost_info/catboost_training.json b/main/train/catboost_info/catboost_training.json index 6c45bd2..0df5170 100644 --- a/main/train/catboost_info/catboost_training.json +++ b/main/train/catboost_info/catboost_training.json @@ -1,792 +1,792 @@ { "meta":{"test_sets":["test"],"test_metrics":[{"best_value":"Min","name":"Logloss"}],"learn_metrics":[{"best_value":"Min","name":"Logloss"}],"launch_mode":"Train","parameters":"","iteration_count":1500,"learn_sets":["learn"],"name":"experiment"}, "iterations":[ -{"learn":[0.6887138297],"iteration":0,"passed_time":0.3283017459,"remaining_time":492.1243171,"test":[0.6894516063]}, -{"learn":[0.6843872854],"iteration":1,"passed_time":0.5412330504,"remaining_time":405.3835548,"test":[0.6855813377]}, -{"learn":[0.6802627852],"iteration":2,"passed_time":0.7661538569,"remaining_time":382.3107746,"test":[0.6821273883]}, -{"learn":[0.6762009319],"iteration":3,"passed_time":0.9717496059,"remaining_time":363.4343526,"test":[0.6785411841]}, -{"learn":[0.6723192545],"iteration":4,"passed_time":1.011827525,"remaining_time":302.5364299,"test":[0.6752505378]}, -{"learn":[0.6682080654],"iteration":5,"passed_time":1.217885731,"remaining_time":303.2535469,"test":[0.6727377765]}, -{"learn":[0.6642342664],"iteration":6,"passed_time":1.441465382,"remaining_time":307.4439736,"test":[0.6695439468]}, -{"learn":[0.66040895],"iteration":7,"passed_time":1.64633775,"remaining_time":307.0419905,"test":[0.6666962384]}, -{"learn":[0.6567665034],"iteration":8,"passed_time":1.849458815,"remaining_time":306.393677,"test":[0.6635423965]}, -{"learn":[0.6530918354],"iteration":9,"passed_time":2.053251668,"remaining_time":305.9344986,"test":[0.660442949]}, -{"learn":[0.6495993504],"iteration":10,"passed_time":2.263635072,"remaining_time":306.4138747,"test":[0.6580995967]}, -{"learn":[0.6459855333],"iteration":11,"passed_time":2.472919541,"remaining_time":306.6420231,"test":[0.6556190011]}, -{"learn":[0.6426157539],"iteration":12,"passed_time":2.67784142,"remaining_time":306.3038609,"test":[0.6533186696]}, -{"learn":[0.6392258492],"iteration":13,"passed_time":2.882790995,"remaining_time":305.9876727,"test":[0.6512931438]}, -{"learn":[0.6358979576],"iteration":14,"passed_time":3.046402744,"remaining_time":301.5938717,"test":[0.6483627862]}, -{"learn":[0.6326337693],"iteration":15,"passed_time":3.264037094,"remaining_time":302.7394404,"test":[0.6462357347]}, -{"learn":[0.6296338495],"iteration":16,"passed_time":3.366605155,"remaining_time":293.6867909,"test":[0.6434692389]}, -{"learn":[0.6265263841],"iteration":17,"passed_time":3.47312979,"remaining_time":285.9543527,"test":[0.6415187001]}, -{"learn":[0.6235141039],"iteration":18,"passed_time":3.687781647,"remaining_time":287.4528747,"test":[0.6387646556]}, -{"learn":[0.6205433947],"iteration":19,"passed_time":3.900772605,"remaining_time":288.6571728,"test":[0.6368737392]}, -{"learn":[0.6175789713],"iteration":20,"passed_time":4.11289922,"remaining_time":289.6656165,"test":[0.63471105]}, -{"learn":[0.6147449654],"iteration":21,"passed_time":4.336435361,"remaining_time":291.329612,"test":[0.6322352833]}, -{"learn":[0.6119898755],"iteration":22,"passed_time":4.568176784,"remaining_time":293.3563961,"test":[0.6300870008]}, -{"learn":[0.6092233761],"iteration":23,"passed_time":4.771516576,"remaining_time":293.4482694,"test":[0.6275533277]}, -{"learn":[0.6065007718],"iteration":24,"passed_time":4.977668345,"remaining_time":293.6824324,"test":[0.6257390345]}, -{"learn":[0.6037280922],"iteration":25,"passed_time":5.183265264,"remaining_time":293.8512692,"test":[0.6242294781]}, -{"learn":[0.601134849],"iteration":26,"passed_time":5.388370956,"remaining_time":293.965571,"test":[0.6220503069]}, -{"learn":[0.5985698655],"iteration":27,"passed_time":5.594962786,"remaining_time":294.1351865,"test":[0.6199216991]}, -{"learn":[0.5959878205],"iteration":28,"passed_time":5.812528469,"remaining_time":294.8354958,"test":[0.6184262302]}, -{"learn":[0.5935677806],"iteration":29,"passed_time":6.030765446,"remaining_time":295.5075069,"test":[0.6162235133]}, -{"learn":[0.5912109686],"iteration":30,"passed_time":6.241235046,"remaining_time":295.7540091,"test":[0.6141742424]}, -{"learn":[0.5888627138],"iteration":31,"passed_time":6.453729887,"remaining_time":296.0648586,"test":[0.612787722]}, -{"learn":[0.5865621045],"iteration":32,"passed_time":6.668549586,"remaining_time":296.4473407,"test":[0.6108688302]}, -{"learn":[0.5842648229],"iteration":33,"passed_time":6.876600376,"remaining_time":296.502828,"test":[0.6096360351]}, -{"learn":[0.5820778865],"iteration":34,"passed_time":7.081681965,"remaining_time":296.4189737,"test":[0.6077157864]}, -{"learn":[0.5799317814],"iteration":35,"passed_time":7.29476106,"remaining_time":296.6536165,"test":[0.6064107181]}, -{"learn":[0.5777438942],"iteration":36,"passed_time":7.524859693,"remaining_time":297.5370198,"test":[0.6055380919]}, -{"learn":[0.5756289014],"iteration":37,"passed_time":7.731563602,"remaining_time":297.4617365,"test":[0.6043700673]}, -{"learn":[0.5736006422],"iteration":38,"passed_time":7.93719022,"remaining_time":297.3393567,"test":[0.6026869514]}, -{"learn":[0.5715183462],"iteration":39,"passed_time":8.151702311,"remaining_time":297.5371343,"test":[0.6015792676]}, -{"learn":[0.5695853779],"iteration":40,"passed_time":8.357713849,"remaining_time":297.412305,"test":[0.5999813949]}, -{"learn":[0.5676576391],"iteration":41,"passed_time":8.570072235,"remaining_time":297.5039362,"test":[0.5986109974]}, -{"learn":[0.5658429393],"iteration":42,"passed_time":8.811404962,"remaining_time":298.5631867,"test":[0.5973064079]}, -{"learn":[0.5640453538],"iteration":43,"passed_time":9.022088624,"remaining_time":298.5491145,"test":[0.59606619]}, -{"learn":[0.562331544],"iteration":44,"passed_time":9.118700355,"remaining_time":294.8379782,"test":[0.594885355]}, -{"learn":[0.5605908479],"iteration":45,"passed_time":9.336381531,"remaining_time":295.1108423,"test":[0.5939087686]}, -{"learn":[0.5587755142],"iteration":46,"passed_time":9.577714971,"remaining_time":296.0940394,"test":[0.5927287318]}, -{"learn":[0.5569407421],"iteration":47,"passed_time":9.787468675,"remaining_time":296.0709274,"test":[0.591221929]}, -{"learn":[0.5552666016],"iteration":48,"passed_time":9.993026113,"remaining_time":295.9159365,"test":[0.590284173]}, -{"learn":[0.5535532672],"iteration":49,"passed_time":10.2041229,"remaining_time":295.9195641,"test":[0.5892440533]}, -{"learn":[0.5520047375],"iteration":50,"passed_time":10.40516421,"remaining_time":295.6290772,"test":[0.5881410786]}, -{"learn":[0.5504471752],"iteration":51,"passed_time":10.61353144,"remaining_time":295.5460292,"test":[0.5866576219]}, -{"learn":[0.5489185594],"iteration":52,"passed_time":10.8183037,"remaining_time":295.3601029,"test":[0.5855571614]}, -{"learn":[0.5474294544],"iteration":53,"passed_time":11.02925288,"remaining_time":295.3388826,"test":[0.5845544753]}, -{"learn":[0.5459806526],"iteration":54,"passed_time":11.23771307,"remaining_time":295.2453706,"test":[0.583309229]}, -{"learn":[0.5445916453],"iteration":55,"passed_time":11.44450368,"remaining_time":295.1047021,"test":[0.5818972075]}, -{"learn":[0.5431495519],"iteration":56,"passed_time":11.6548216,"remaining_time":295.0510099,"test":[0.580668882]}, -{"learn":[0.5417205055],"iteration":57,"passed_time":11.86014598,"remaining_time":294.8677672,"test":[0.5798167684]}, -{"learn":[0.5403851653],"iteration":58,"passed_time":12.06903795,"remaining_time":294.77091,"test":[0.5785289403]}, -{"learn":[0.5389834808],"iteration":59,"passed_time":12.28734105,"remaining_time":294.8961853,"test":[0.5776724369]}, -{"learn":[0.5377107875],"iteration":60,"passed_time":12.504178,"remaining_time":294.9756088,"test":[0.5765606026]}, -{"learn":[0.5364798765],"iteration":61,"passed_time":12.77687786,"remaining_time":296.341135,"test":[0.5753093701]}, -{"learn":[0.5352338583],"iteration":62,"passed_time":13.01008928,"remaining_time":296.7539412,"test":[0.5742001698]}, -{"learn":[0.5339945486],"iteration":63,"passed_time":13.21914713,"remaining_time":296.6046137,"test":[0.5732021129]}, -{"learn":[0.5327745188],"iteration":64,"passed_time":13.42735441,"remaining_time":296.4346705,"test":[0.5721310646]}, -{"learn":[0.5315706526],"iteration":65,"passed_time":13.63547052,"remaining_time":296.2615867,"test":[0.5711689648]}, -{"learn":[0.5304691555],"iteration":66,"passed_time":13.87096536,"remaining_time":296.6730352,"test":[0.570040489]}, -{"learn":[0.5293499101],"iteration":67,"passed_time":14.07559059,"remaining_time":296.4153782,"test":[0.5690538059]}, -{"learn":[0.528244557],"iteration":68,"passed_time":14.28294464,"remaining_time":296.2158518,"test":[0.5685098612]}, -{"learn":[0.5271158037],"iteration":69,"passed_time":14.48864803,"remaining_time":295.9823813,"test":[0.567835218]}, -{"learn":[0.5260443622],"iteration":70,"passed_time":14.69267194,"remaining_time":295.7158902,"test":[0.5669511382]}, -{"learn":[0.5249990677],"iteration":71,"passed_time":14.90912476,"remaining_time":295.6976411,"test":[0.5659244275]}, -{"learn":[0.5239809237],"iteration":72,"passed_time":15.11450581,"remaining_time":295.4575314,"test":[0.5654294897]}, -{"learn":[0.5229378476],"iteration":73,"passed_time":15.32745061,"remaining_time":295.3641159,"test":[0.5649020668]}, -{"learn":[0.5219464315],"iteration":74,"passed_time":15.53820445,"remaining_time":295.2258846,"test":[0.564184363]}, -{"learn":[0.5208603056],"iteration":75,"passed_time":15.75581809,"remaining_time":295.2142757,"test":[0.5636825211]}, -{"learn":[0.5199106717],"iteration":76,"passed_time":15.96092236,"remaining_time":294.9661366,"test":[0.5628025917]}, -{"learn":[0.5189214741],"iteration":77,"passed_time":16.17127004,"remaining_time":294.8146924,"test":[0.5626036907]}, -{"learn":[0.5179007418],"iteration":78,"passed_time":16.3854461,"remaining_time":294.730619,"test":[0.5622698611]}, -{"learn":[0.5170257982],"iteration":79,"passed_time":16.59669052,"remaining_time":294.5912567,"test":[0.5612862909]}, -{"learn":[0.5160979269],"iteration":80,"passed_time":16.81674878,"remaining_time":294.6045248,"test":[0.5607771857]}, -{"learn":[0.5152054464],"iteration":81,"passed_time":17.08411486,"remaining_time":295.4301814,"test":[0.5598296124]}, -{"learn":[0.5143571779],"iteration":82,"passed_time":17.2847474,"remaining_time":295.0902057,"test":[0.5594330475]}, -{"learn":[0.5135479977],"iteration":83,"passed_time":17.49421614,"remaining_time":294.9025007,"test":[0.5586549232]}, -{"learn":[0.5127352784],"iteration":84,"passed_time":17.70076062,"remaining_time":294.6656033,"test":[0.5578689769]}, -{"learn":[0.511893507],"iteration":85,"passed_time":17.90903378,"remaining_time":294.4578344,"test":[0.5573709662]}, -{"learn":[0.5110974267],"iteration":86,"passed_time":18.12669175,"remaining_time":294.4024763,"test":[0.5569637459]}, -{"learn":[0.5103353109],"iteration":87,"passed_time":18.33935008,"remaining_time":294.263208,"test":[0.5566546991]}, -{"learn":[0.5094643818],"iteration":88,"passed_time":18.60457541,"remaining_time":294.9556843,"test":[0.5559996507]}, -{"learn":[0.5086953991],"iteration":89,"passed_time":18.82047814,"remaining_time":294.8541575,"test":[0.5555009615]}, -{"learn":[0.5079769143],"iteration":90,"passed_time":19.03668631,"remaining_time":294.7548463,"test":[0.5550565471]}, -{"learn":[0.5072723744],"iteration":91,"passed_time":19.2409811,"remaining_time":294.4706673,"test":[0.5546474911]}, -{"learn":[0.5064938838],"iteration":92,"passed_time":19.45755733,"remaining_time":294.3740124,"test":[0.5539547297]}, -{"learn":[0.5056995994],"iteration":93,"passed_time":19.66756889,"remaining_time":294.1766156,"test":[0.5533245839]}, -{"learn":[0.5049644228],"iteration":94,"passed_time":19.87189313,"remaining_time":293.8948406,"test":[0.5528296062]}, -{"learn":[0.5042152484],"iteration":95,"passed_time":20.07848777,"remaining_time":293.6478837,"test":[0.5525477765]}, -{"learn":[0.5035045284],"iteration":96,"passed_time":20.28909943,"remaining_time":293.4598609,"test":[0.5519693514]}, -{"learn":[0.5028365414],"iteration":97,"passed_time":20.49512857,"remaining_time":293.205819,"test":[0.5516401111]}, -{"learn":[0.5021301528],"iteration":98,"passed_time":20.70482965,"remaining_time":293.0047105,"test":[0.5510558195]}, -{"learn":[0.5014404559],"iteration":99,"passed_time":20.91493833,"remaining_time":292.8091366,"test":[0.5507846453]}, -{"learn":[0.500777487],"iteration":100,"passed_time":21.14971809,"remaining_time":292.955006,"test":[0.5506432915]}, -{"learn":[0.5001429892],"iteration":101,"passed_time":21.36933356,"remaining_time":292.8855717,"test":[0.5502156384]}, -{"learn":[0.4995491114],"iteration":102,"passed_time":21.58716298,"remaining_time":292.7889969,"test":[0.5496516599]}, -{"learn":[0.4989485253],"iteration":103,"passed_time":21.81016169,"remaining_time":292.7594781,"test":[0.5492036537]}, -{"learn":[0.4983499992],"iteration":104,"passed_time":22.01813865,"remaining_time":292.5266992,"test":[0.5488685869]}, -{"learn":[0.4977427575],"iteration":105,"passed_time":22.2309777,"remaining_time":292.3583294,"test":[0.5482614281]}, -{"learn":[0.4971949405],"iteration":106,"passed_time":22.43811073,"remaining_time":292.1148435,"test":[0.5479093605]}, -{"learn":[0.4966445881],"iteration":107,"passed_time":22.64579067,"remaining_time":291.8790797,"test":[0.5476357519]}, -{"learn":[0.4960738992],"iteration":108,"passed_time":22.8561471,"remaining_time":291.6779874,"test":[0.5473086268]}, -{"learn":[0.4954706192],"iteration":109,"passed_time":23.06363527,"remaining_time":291.4404821,"test":[0.5471436475]}, -{"learn":[0.4948638001],"iteration":110,"passed_time":23.27268242,"remaining_time":291.2230259,"test":[0.5466196967]}, -{"learn":[0.4943255439],"iteration":111,"passed_time":23.48335216,"remaining_time":291.0258285,"test":[0.5463429353]}, -{"learn":[0.4937688],"iteration":112,"passed_time":23.69308683,"remaining_time":290.8169154,"test":[0.5462688264]}, -{"learn":[0.4932566907],"iteration":113,"passed_time":23.90256435,"remaining_time":290.6048613,"test":[0.5460389967]}, -{"learn":[0.4927494411],"iteration":114,"passed_time":24.10820629,"remaining_time":290.3466584,"test":[0.5458030612]}, -{"learn":[0.4922379656],"iteration":115,"passed_time":24.31439431,"remaining_time":290.0958769,"test":[0.5455376337]}, -{"learn":[0.4917602435],"iteration":116,"passed_time":24.55630218,"remaining_time":290.2680848,"test":[0.5450453697]}, -{"learn":[0.4912688932],"iteration":117,"passed_time":24.80212849,"remaining_time":290.4791659,"test":[0.5448329]}, -{"learn":[0.4907662919],"iteration":118,"passed_time":25.01110113,"remaining_time":290.2548795,"test":[0.5444673436]}, -{"learn":[0.4903079026],"iteration":119,"passed_time":25.22473218,"remaining_time":290.0844201,"test":[0.5442106959]}, -{"learn":[0.4896477861],"iteration":120,"passed_time":25.48816229,"remaining_time":290.4807917,"test":[0.5439808263]}, -{"learn":[0.489197056],"iteration":121,"passed_time":25.72516541,"remaining_time":290.567852,"test":[0.5434537626]}, -{"learn":[0.4886972542],"iteration":122,"passed_time":25.9664879,"remaining_time":290.6979987,"test":[0.5430699683]}, -{"learn":[0.4882628989],"iteration":123,"passed_time":26.17355547,"remaining_time":290.4420349,"test":[0.5427608018]}, -{"learn":[0.4878229445],"iteration":124,"passed_time":26.38205433,"remaining_time":290.2025976,"test":[0.5426135416]}, -{"learn":[0.4874311637],"iteration":125,"passed_time":26.59233696,"remaining_time":289.983103,"test":[0.5422171364]}, -{"learn":[0.4870381152],"iteration":126,"passed_time":26.80138773,"remaining_time":289.7504358,"test":[0.5420255385]}, -{"learn":[0.4866496622],"iteration":127,"passed_time":27.01016614,"remaining_time":289.5152183,"test":[0.5418352975]}, -{"learn":[0.4862709813],"iteration":128,"passed_time":27.21670162,"remaining_time":289.256573,"test":[0.5415267296]}, -{"learn":[0.4858681607],"iteration":129,"passed_time":27.44640729,"remaining_time":289.2429076,"test":[0.5410198194]}, -{"learn":[0.4853899103],"iteration":130,"passed_time":27.68396245,"remaining_time":289.3079741,"test":[0.5406753343]}, -{"learn":[0.4849540232],"iteration":131,"passed_time":27.90148286,"remaining_time":289.1608224,"test":[0.5404109844]}, -{"learn":[0.4845096845],"iteration":132,"passed_time":28.11763031,"remaining_time":288.998501,"test":[0.5400712883]}, -{"learn":[0.4841039059],"iteration":133,"passed_time":28.32493981,"remaining_time":288.7452819,"test":[0.5399532407]}, -{"learn":[0.4836538097],"iteration":134,"passed_time":28.53715616,"remaining_time":288.5423567,"test":[0.539657563]}, -{"learn":[0.4832327127],"iteration":135,"passed_time":28.7517187,"remaining_time":288.3628258,"test":[0.539344725]}, -{"learn":[0.4828748437],"iteration":136,"passed_time":28.96152401,"remaining_time":288.1354543,"test":[0.5392009767]}, -{"learn":[0.4824942083],"iteration":137,"passed_time":29.17156026,"remaining_time":287.9106165,"test":[0.5390713558]}, -{"learn":[0.4820718964],"iteration":138,"passed_time":29.37967995,"remaining_time":287.667226,"test":[0.5387477027]}, -{"learn":[0.4817320925],"iteration":139,"passed_time":29.65072058,"remaining_time":288.0355714,"test":[0.5385680373]}, -{"learn":[0.4813566338],"iteration":140,"passed_time":29.86537745,"remaining_time":287.851404,"test":[0.5384274019]}, -{"learn":[0.48099718],"iteration":141,"passed_time":30.08600662,"remaining_time":287.7239224,"test":[0.5381120496]}, -{"learn":[0.4806665143],"iteration":142,"passed_time":30.30257312,"remaining_time":287.5565855,"test":[0.5377796167]}, -{"learn":[0.480273413],"iteration":143,"passed_time":30.50726137,"remaining_time":287.2767113,"test":[0.5376424134]}, -{"learn":[0.4799522024],"iteration":144,"passed_time":30.73843311,"remaining_time":287.2453577,"test":[0.5373759882]}, -{"learn":[0.4796210085],"iteration":145,"passed_time":30.96263978,"remaining_time":287.146673,"test":[0.5372649644]}, -{"learn":[0.4792147017],"iteration":146,"passed_time":31.17488357,"remaining_time":286.9361733,"test":[0.5369387173]}, -{"learn":[0.4789183174],"iteration":147,"passed_time":31.38011359,"remaining_time":286.6615782,"test":[0.536795847]}, -{"learn":[0.478532664],"iteration":148,"passed_time":31.59355334,"remaining_time":286.4623528,"test":[0.5366868585]}, -{"learn":[0.4782636944],"iteration":149,"passed_time":31.7972291,"remaining_time":286.1750619,"test":[0.5365283044]}, -{"learn":[0.4779824172],"iteration":150,"passed_time":32.00218754,"remaining_time":285.9003377,"test":[0.5363033834]}, -{"learn":[0.4776756799],"iteration":151,"passed_time":32.21040916,"remaining_time":285.6554707,"test":[0.5361123842]}, -{"learn":[0.4773818311],"iteration":152,"passed_time":32.41875052,"remaining_time":285.412137,"test":[0.5359008324]}, -{"learn":[0.4770647406],"iteration":153,"passed_time":32.63514375,"remaining_time":285.239633,"test":[0.5357652652]}, -{"learn":[0.4767062377],"iteration":154,"passed_time":32.84811293,"remaining_time":285.0368509,"test":[0.5357864164]}, -{"learn":[0.4764296089],"iteration":155,"passed_time":33.0567465,"remaining_time":284.7965852,"test":[0.535719012]}, -{"learn":[0.4760174917],"iteration":156,"passed_time":33.27242701,"remaining_time":284.617003,"test":[0.5356127372]}, -{"learn":[0.4756622109],"iteration":157,"passed_time":33.49206585,"remaining_time":284.4705846,"test":[0.5353806727]}, -{"learn":[0.4753690488],"iteration":158,"passed_time":33.76055951,"remaining_time":284.7352849,"test":[0.535390969]}, -{"learn":[0.4751120698],"iteration":159,"passed_time":33.97791398,"remaining_time":284.5650296,"test":[0.5351500849]}, -{"learn":[0.4748108259],"iteration":160,"passed_time":34.24523877,"remaining_time":284.8097809,"test":[0.5351811332]}, -{"learn":[0.4745126986],"iteration":161,"passed_time":34.45225547,"remaining_time":284.55011,"test":[0.5351523197]}, -{"learn":[0.4742214381],"iteration":162,"passed_time":34.69581548,"remaining_time":284.59083,"test":[0.5348433527]}, -{"learn":[0.4739676812],"iteration":163,"passed_time":34.90263593,"remaining_time":284.3287902,"test":[0.5345701033]}, -{"learn":[0.4736240741],"iteration":164,"passed_time":35.11130664,"remaining_time":284.0823901,"test":[0.5343011639]}, -{"learn":[0.4733877484],"iteration":165,"passed_time":35.31858925,"remaining_time":283.8252895,"test":[0.5340034509]}, -{"learn":[0.4731457709],"iteration":166,"passed_time":35.5215187,"remaining_time":283.5340385,"test":[0.5338239053]}, -{"learn":[0.4729062759],"iteration":167,"passed_time":35.72664687,"remaining_time":283.2612716,"test":[0.533652421]}, -{"learn":[0.4726252629],"iteration":168,"passed_time":35.93364178,"remaining_time":283.0040072,"test":[0.5334977778]}, -{"learn":[0.4723725624],"iteration":169,"passed_time":36.14603772,"remaining_time":282.7895892,"test":[0.5333416182]}, -{"learn":[0.4720965675],"iteration":170,"passed_time":36.35991691,"remaining_time":282.5867226,"test":[0.5331617932]}, -{"learn":[0.4718570197],"iteration":171,"passed_time":36.57295352,"remaining_time":282.3772225,"test":[0.5330645775]}, -{"learn":[0.4715016333],"iteration":172,"passed_time":36.81022333,"remaining_time":282.3535628,"test":[0.5329320833]}, -{"learn":[0.4712788301],"iteration":173,"passed_time":37.02094352,"remaining_time":282.1251213,"test":[0.5327248415]}, -{"learn":[0.4709921123],"iteration":174,"passed_time":37.23646649,"remaining_time":281.9332463,"test":[0.532655721]}, -{"learn":[0.4707643438],"iteration":175,"passed_time":37.45219574,"remaining_time":281.7426543,"test":[0.5324228584]}, -{"learn":[0.4704056825],"iteration":176,"passed_time":37.6691987,"remaining_time":281.5612987,"test":[0.5324790887]}, -{"learn":[0.4701234017],"iteration":177,"passed_time":37.90491523,"remaining_time":281.5185277,"test":[0.5322301032]}, -{"learn":[0.4698584465],"iteration":178,"passed_time":38.16613904,"remaining_time":281.6618418,"test":[0.5321801385]}, -{"learn":[0.4695728908],"iteration":179,"passed_time":38.37733401,"remaining_time":281.4337827,"test":[0.5319024992]}, -{"learn":[0.4693260536],"iteration":180,"passed_time":38.6330386,"remaining_time":281.5302647,"test":[0.531768249]}, -{"learn":[0.4690976513],"iteration":181,"passed_time":38.84879856,"remaining_time":281.3336072,"test":[0.5317025606]}, -{"learn":[0.4688616426],"iteration":182,"passed_time":39.05676562,"remaining_time":281.0806575,"test":[0.531470576]}, -{"learn":[0.4685305015],"iteration":183,"passed_time":39.27704777,"remaining_time":280.9162765,"test":[0.5313124609]}, -{"learn":[0.4682865695],"iteration":184,"passed_time":39.50365307,"remaining_time":280.7962367,"test":[0.5312031132]}, -{"learn":[0.4680025456],"iteration":185,"passed_time":39.72294934,"remaining_time":280.6234163,"test":[0.5313067939]}, -{"learn":[0.4677844436],"iteration":186,"passed_time":39.93898545,"remaining_time":280.427208,"test":[0.5310851454]}, -{"learn":[0.4674939226],"iteration":187,"passed_time":40.15035349,"remaining_time":280.1982116,"test":[0.5309420357]}, -{"learn":[0.4672482474],"iteration":188,"passed_time":40.36996573,"remaining_time":280.0265877,"test":[0.5308133726]}, -{"learn":[0.4669966563],"iteration":189,"passed_time":40.57890753,"remaining_time":279.7808887,"test":[0.5305983089]}, -{"learn":[0.4666803581],"iteration":190,"passed_time":40.80276763,"remaining_time":279.6378159,"test":[0.5304503703]}, -{"learn":[0.4664588227],"iteration":191,"passed_time":41.00915922,"remaining_time":279.3748972,"test":[0.5304046358]}, -{"learn":[0.4662119855],"iteration":192,"passed_time":41.21768415,"remaining_time":279.1270113,"test":[0.5304474969]}, -{"learn":[0.4660336056],"iteration":193,"passed_time":41.42537817,"remaining_time":278.8739376,"test":[0.5302670334]}, -{"learn":[0.465759565],"iteration":194,"passed_time":41.63552693,"remaining_time":278.6377571,"test":[0.530215632]}, -{"learn":[0.4655335925],"iteration":195,"passed_time":41.84387896,"remaining_time":278.3898886,"test":[0.5302616857]}, -{"learn":[0.4653787184],"iteration":196,"passed_time":42.07151485,"remaining_time":278.2699688,"test":[0.5301574862]}, -{"learn":[0.4651595599],"iteration":197,"passed_time":42.31569189,"remaining_time":278.2577315,"test":[0.5299640126]}, -{"learn":[0.4649592589],"iteration":198,"passed_time":42.52817255,"remaining_time":278.0359421,"test":[0.5298924179]}, -{"learn":[0.4647202922],"iteration":199,"passed_time":42.75850677,"remaining_time":277.930294,"test":[0.5299083411]}, -{"learn":[0.4645238472],"iteration":200,"passed_time":42.99071311,"remaining_time":277.8355041,"test":[0.5297052099]}, -{"learn":[0.4642532401],"iteration":201,"passed_time":43.20050556,"remaining_time":277.5953278,"test":[0.5296109474]}, -{"learn":[0.4640300671],"iteration":202,"passed_time":43.41431105,"remaining_time":277.3810908,"test":[0.5295602644]}, -{"learn":[0.4638038305],"iteration":203,"passed_time":43.62268492,"remaining_time":277.1323513,"test":[0.5294535905]}, -{"learn":[0.4636120338],"iteration":204,"passed_time":43.88510475,"remaining_time":277.2254178,"test":[0.5293538607]}, -{"learn":[0.4633936149],"iteration":205,"passed_time":44.09203525,"remaining_time":276.9664738,"test":[0.5292528139]}, -{"learn":[0.4631845454],"iteration":206,"passed_time":44.2987429,"remaining_time":276.7066404,"test":[0.5291430272]}, -{"learn":[0.4630175222],"iteration":207,"passed_time":44.50942467,"remaining_time":276.4720033,"test":[0.5291064317]}, -{"learn":[0.4627118941],"iteration":208,"passed_time":44.71988676,"remaining_time":276.2362383,"test":[0.5291368016]}, -{"learn":[0.462500342],"iteration":209,"passed_time":44.93071709,"remaining_time":276.0029764,"test":[0.5290519175]}, -{"learn":[0.4623026821],"iteration":210,"passed_time":45.14892317,"remaining_time":275.8149856,"test":[0.5290010349]}, -{"learn":[0.4620509325],"iteration":211,"passed_time":45.35339107,"remaining_time":275.5432439,"test":[0.528878797]}, -{"learn":[0.4618490996],"iteration":212,"passed_time":45.56580477,"remaining_time":275.3201443,"test":[0.5288099957]}, -{"learn":[0.4616663883],"iteration":213,"passed_time":45.80824569,"remaining_time":275.2775886,"test":[0.5287029228]}, -{"learn":[0.4614388839],"iteration":214,"passed_time":46.02760885,"remaining_time":275.0952436,"test":[0.5286111347]}, -{"learn":[0.4611913601],"iteration":215,"passed_time":46.30318522,"remaining_time":275.2467122,"test":[0.5286519604]}, -{"learn":[0.4609538723],"iteration":216,"passed_time":46.53420502,"remaining_time":275.1308066,"test":[0.5285690318]}, -{"learn":[0.4608258318],"iteration":217,"passed_time":46.74960881,"remaining_time":274.9220115,"test":[0.5284838683]}, -{"learn":[0.4606162869],"iteration":218,"passed_time":46.97150736,"remaining_time":274.7511458,"test":[0.5283307018]}, -{"learn":[0.4604787385],"iteration":219,"passed_time":47.17985731,"remaining_time":274.500988,"test":[0.5283035644]}, -{"learn":[0.4602659714],"iteration":220,"passed_time":47.38620008,"remaining_time":274.2395923,"test":[0.5281441723]}, -{"learn":[0.4600531516],"iteration":221,"passed_time":47.60357505,"remaining_time":274.0422023,"test":[0.5279370103]}, -{"learn":[0.4598740322],"iteration":222,"passed_time":47.81945953,"remaining_time":273.8360978,"test":[0.5279410809]}, -{"learn":[0.4596930112],"iteration":223,"passed_time":48.03154738,"remaining_time":273.6082788,"test":[0.5277739865]}, -{"learn":[0.4594828853],"iteration":224,"passed_time":48.25679383,"remaining_time":273.455165,"test":[0.5278144531]}, -{"learn":[0.4593606024],"iteration":225,"passed_time":48.47838951,"remaining_time":273.2808329,"test":[0.5277773787]}, -{"learn":[0.4590814381],"iteration":226,"passed_time":48.70127068,"remaining_time":273.1132933,"test":[0.5276907386]}, -{"learn":[0.4589115626],"iteration":227,"passed_time":48.92925623,"remaining_time":272.9737453,"test":[0.5275788768]}, -{"learn":[0.458786797],"iteration":228,"passed_time":49.13740749,"remaining_time":272.7233403,"test":[0.5274718039]}, -{"learn":[0.4586156537],"iteration":229,"passed_time":49.34267414,"remaining_time":272.4573746,"test":[0.5273879175]}, -{"learn":[0.4584218498],"iteration":230,"passed_time":49.55847402,"remaining_time":272.2497989,"test":[0.5272032237]}, -{"learn":[0.4582468505],"iteration":231,"passed_time":49.7683747,"remaining_time":272.00991,"test":[0.5271434416]}, -{"learn":[0.4580919236],"iteration":232,"passed_time":49.97549033,"remaining_time":271.7551341,"test":[0.5271729336]}, -{"learn":[0.4579602912],"iteration":233,"passed_time":50.20951486,"remaining_time":271.6463496,"test":[0.5271931669]}, -{"learn":[0.4577697094],"iteration":234,"passed_time":50.45625736,"remaining_time":271.6049598,"test":[0.5271494278]}, -{"learn":[0.4575785994],"iteration":235,"passed_time":50.72549762,"remaining_time":271.6823262,"test":[0.5270163749]}, -{"learn":[0.4574154851],"iteration":236,"passed_time":50.93444754,"remaining_time":271.4354736,"test":[0.5268986864]}, -{"learn":[0.4572040914],"iteration":237,"passed_time":51.15135621,"remaining_time":271.2311409,"test":[0.5268795306]}, -{"learn":[0.4570666486],"iteration":238,"passed_time":51.36916907,"remaining_time":271.0314737,"test":[0.5269486112]}, -{"learn":[0.4568047043],"iteration":239,"passed_time":51.59087682,"remaining_time":270.8521033,"test":[0.5269823733]}, -{"learn":[0.4566408504],"iteration":240,"passed_time":51.82182656,"remaining_time":270.7206624,"test":[0.5268644055]}, -{"learn":[0.456427608],"iteration":241,"passed_time":52.04138441,"remaining_time":270.5291801,"test":[0.5268694739]}, -{"learn":[0.4562375017],"iteration":242,"passed_time":52.24611907,"remaining_time":270.2607888,"test":[0.5267796811]}, -{"learn":[0.4559405892],"iteration":243,"passed_time":52.46606132,"remaining_time":270.0712009,"test":[0.5267057317]}, -{"learn":[0.4558015617],"iteration":244,"passed_time":52.6872576,"remaining_time":269.887789,"test":[0.5267355429]}, -{"learn":[0.4556457896],"iteration":245,"passed_time":52.8948513,"remaining_time":269.6347298,"test":[0.5266087954]}, -{"learn":[0.4554681492],"iteration":246,"passed_time":53.12432641,"remaining_time":269.4930404,"test":[0.5265850901]}, -{"learn":[0.4552844343],"iteration":247,"passed_time":53.34081813,"remaining_time":269.285098,"test":[0.5266133449]}, -{"learn":[0.4549960262],"iteration":248,"passed_time":53.5555288,"remaining_time":269.0681387,"test":[0.5264907079]}, -{"learn":[0.4547091499],"iteration":249,"passed_time":53.7669082,"remaining_time":268.834541,"test":[0.526436513]}, -{"learn":[0.4544524878],"iteration":250,"passed_time":53.9749097,"remaining_time":268.5843116,"test":[0.526419153]}, -{"learn":[0.4542734741],"iteration":251,"passed_time":54.18101643,"remaining_time":268.3250338,"test":[0.5263015844]}, -{"learn":[0.4541751195],"iteration":252,"passed_time":54.38888617,"remaining_time":268.0748658,"test":[0.5262358959]}, -{"learn":[0.4540318662],"iteration":253,"passed_time":54.60797708,"remaining_time":267.8800765,"test":[0.526113259]}, -{"learn":[0.4538242758],"iteration":254,"passed_time":54.84165749,"remaining_time":267.7563277,"test":[0.5261479788]}, -{"learn":[0.4536671832],"iteration":255,"passed_time":55.07439841,"remaining_time":267.6271548,"test":[0.5260264992]}, -{"learn":[0.4534657729],"iteration":256,"passed_time":55.28319647,"remaining_time":267.3813743,"test":[0.5259905821]}, -{"learn":[0.4533309183],"iteration":257,"passed_time":55.5104133,"remaining_time":267.2245477,"test":[0.5258837885]}, -{"learn":[0.4531104921],"iteration":258,"passed_time":55.72899607,"remaining_time":267.0258074,"test":[0.5257997824]}, -{"learn":[0.4528481781],"iteration":259,"passed_time":55.95400479,"remaining_time":266.8575613,"test":[0.525747064]}, -{"learn":[0.4526748163],"iteration":260,"passed_time":56.17438366,"remaining_time":266.6669018,"test":[0.5256947847]}, -{"learn":[0.4525394863],"iteration":261,"passed_time":56.38599992,"remaining_time":266.4346103,"test":[0.5257121047]}, -{"learn":[0.4523340087],"iteration":262,"passed_time":56.59841305,"remaining_time":266.2062241,"test":[0.5255856764]}, -{"learn":[0.452107508],"iteration":263,"passed_time":56.81347404,"remaining_time":265.9903557,"test":[0.5255532712]}, -{"learn":[0.4519560673],"iteration":264,"passed_time":57.02882367,"remaining_time":265.7758386,"test":[0.5254442029]}, -{"learn":[0.4517016766],"iteration":265,"passed_time":57.24719256,"remaining_time":265.5753219,"test":[0.5253342167]}, -{"learn":[0.451566241],"iteration":266,"passed_time":57.46008151,"remaining_time":265.3493652,"test":[0.5252585913]}, -{"learn":[0.4513973162],"iteration":267,"passed_time":57.67532472,"remaining_time":265.1343286,"test":[0.5251232236]}, -{"learn":[0.4511307236],"iteration":268,"passed_time":57.90495491,"remaining_time":264.9851282,"test":[0.5250109228]}, -{"learn":[0.4510252909],"iteration":269,"passed_time":58.1474584,"remaining_time":264.8939772,"test":[0.5251278929]}, -{"learn":[0.4507913422],"iteration":270,"passed_time":58.35251256,"remaining_time":264.6318743,"test":[0.5251599389]}, -{"learn":[0.4506708024],"iteration":271,"passed_time":58.55642793,"remaining_time":264.3650496,"test":[0.5251249397]}, -{"learn":[0.4504389667],"iteration":272,"passed_time":58.76797556,"remaining_time":264.1329891,"test":[0.5250323134]}, -{"learn":[0.4502753241],"iteration":273,"passed_time":59.04319493,"remaining_time":264.1859744,"test":[0.5249261584]}, -{"learn":[0.4500877004],"iteration":274,"passed_time":59.32124598,"remaining_time":264.2491866,"test":[0.524854404]}, -{"learn":[0.44989284],"iteration":275,"passed_time":59.53042552,"remaining_time":264.0044958,"test":[0.5247326849]}, -{"learn":[0.4497752582],"iteration":276,"passed_time":59.73013362,"remaining_time":263.7182434,"test":[0.5246913403]}, -{"learn":[0.4497071707],"iteration":277,"passed_time":59.93395611,"remaining_time":263.4506991,"test":[0.5246368261]}, -{"learn":[0.4496092387],"iteration":278,"passed_time":60.13539014,"remaining_time":263.173159,"test":[0.52460047]}, -{"learn":[0.4494378841],"iteration":279,"passed_time":60.34248402,"remaining_time":262.9208232,"test":[0.5244898054]}, -{"learn":[0.4490959673],"iteration":280,"passed_time":60.57009365,"remaining_time":262.7578084,"test":[0.5244507355]}, -{"learn":[0.4489082908],"iteration":281,"passed_time":60.77504949,"remaining_time":262.4964903,"test":[0.5244589566]}, -{"learn":[0.4487697387],"iteration":282,"passed_time":60.99505059,"remaining_time":262.3002706,"test":[0.5244354508]}, -{"learn":[0.448658654],"iteration":283,"passed_time":61.19978703,"remaining_time":262.0385248,"test":[0.5244328568]}, -{"learn":[0.4484694984],"iteration":284,"passed_time":61.4067821,"remaining_time":261.7868079,"test":[0.5243957424]}, -{"learn":[0.4483210158],"iteration":285,"passed_time":61.61940357,"remaining_time":261.5592865,"test":[0.5243398714]}, -{"learn":[0.4481748573],"iteration":286,"passed_time":61.82488658,"remaining_time":261.3016984,"test":[0.5242596964]}, -{"learn":[0.4479443949],"iteration":287,"passed_time":62.03413996,"remaining_time":261.060339,"test":[0.5241995552]}, -{"learn":[0.4477727762],"iteration":288,"passed_time":62.241505,"remaining_time":260.8112891,"test":[0.5241163073]}, -{"learn":[0.4476346467],"iteration":289,"passed_time":62.44314116,"remaining_time":260.5386235,"test":[0.5240827048]}, -{"learn":[0.4475678269],"iteration":290,"passed_time":62.6413763,"remaining_time":260.252316,"test":[0.5240546496]}, -{"learn":[0.4474215628],"iteration":291,"passed_time":62.85320449,"remaining_time":260.022846,"test":[0.5239532436]}, -{"learn":[0.4472752458],"iteration":292,"passed_time":63.08103516,"remaining_time":259.8594179,"test":[0.5238582229]}, -{"learn":[0.4470165765],"iteration":293,"passed_time":63.34423992,"remaining_time":259.8406576,"test":[0.5238159604]}, -{"learn":[0.4468634455],"iteration":294,"passed_time":63.55086079,"remaining_time":259.5891093,"test":[0.5237865882]}, -{"learn":[0.4466839564],"iteration":295,"passed_time":63.76301977,"remaining_time":259.3603912,"test":[0.5237659159]}, -{"learn":[0.4465560743],"iteration":296,"passed_time":63.97200053,"remaining_time":259.1189112,"test":[0.5237495536]}, -{"learn":[0.4463076524],"iteration":297,"passed_time":64.19045345,"remaining_time":258.9158559,"test":[0.5237665943]}, -{"learn":[0.4461113659],"iteration":298,"passed_time":64.39927163,"remaining_time":258.6739974,"test":[0.5237160709]}, -{"learn":[0.4460250547],"iteration":299,"passed_time":64.59942912,"remaining_time":258.3977165,"test":[0.5237108829]}, -{"learn":[0.4458235388],"iteration":300,"passed_time":64.8091885,"remaining_time":258.1601894,"test":[0.5236443165]}, -{"learn":[0.4456807081],"iteration":301,"passed_time":65.01937985,"remaining_time":257.9245598,"test":[0.5236073618]}, -{"learn":[0.4454037096],"iteration":302,"passed_time":65.22134624,"remaining_time":257.6566054,"test":[0.5235901215]}, -{"learn":[0.4453104787],"iteration":303,"passed_time":65.4254174,"remaining_time":257.3973658,"test":[0.5235915183]}, -{"learn":[0.4451840757],"iteration":304,"passed_time":65.64121934,"remaining_time":257.1844496,"test":[0.5234947816]}, -{"learn":[0.4450407696],"iteration":305,"passed_time":65.84752166,"remaining_time":256.9344473,"test":[0.5234583057]}, -{"learn":[0.4449429961],"iteration":306,"passed_time":66.05035229,"remaining_time":256.6712387,"test":[0.5234427417]}, -{"learn":[0.4447875938],"iteration":307,"passed_time":66.26262236,"remaining_time":256.4449541,"test":[0.5233568598]}, -{"learn":[0.4447108434],"iteration":308,"passed_time":66.46738856,"remaining_time":256.1898375,"test":[0.5233460847]}, -{"learn":[0.4446476155],"iteration":309,"passed_time":66.67338488,"remaining_time":255.9397678,"test":[0.5232778023]}, -{"learn":[0.4445679071],"iteration":310,"passed_time":66.87576787,"remaining_time":255.6761672,"test":[0.5232615997]}, -{"learn":[0.4443813398],"iteration":311,"passed_time":67.11817838,"remaining_time":255.5653715,"test":[0.5232866219]}, -{"learn":[0.4442457457],"iteration":312,"passed_time":67.35370179,"remaining_time":255.4276167,"test":[0.5232186189]}, -{"learn":[0.4441295373],"iteration":313,"passed_time":67.62310596,"remaining_time":255.4172091,"test":[0.5232710179]}, -{"learn":[0.4439701205],"iteration":314,"passed_time":67.85493005,"remaining_time":255.2637845,"test":[0.52324412]}, -{"learn":[0.4436869418],"iteration":315,"passed_time":68.06792468,"remaining_time":255.0393127,"test":[0.5232147079]}, -{"learn":[0.4436074446],"iteration":316,"passed_time":68.27243536,"remaining_time":254.7832524,"test":[0.523215945]}, -{"learn":[0.4433313969],"iteration":317,"passed_time":68.49371266,"remaining_time":254.5898376,"test":[0.5232299128]}, -{"learn":[0.4431709236],"iteration":318,"passed_time":68.70155234,"remaining_time":254.3464994,"test":[0.5231811453]}, -{"learn":[0.4430649099],"iteration":319,"passed_time":68.90937107,"remaining_time":254.1033058,"test":[0.5232248046]}, -{"learn":[0.4429951056],"iteration":320,"passed_time":69.11363136,"remaining_time":253.8472628,"test":[0.5231860141]}, -{"learn":[0.4428992073],"iteration":321,"passed_time":69.31792761,"remaining_time":253.5916731,"test":[0.5232447984]}, -{"learn":[0.4428329157],"iteration":322,"passed_time":69.52151745,"remaining_time":253.3338267,"test":[0.5232055291]}, -{"learn":[0.4427575123],"iteration":323,"passed_time":69.72620267,"remaining_time":253.0802912,"test":[0.5232274784]}, -{"learn":[0.4426940995],"iteration":324,"passed_time":69.93449947,"remaining_time":252.8401135,"test":[0.5231534093]}, -{"learn":[0.4426260119],"iteration":325,"passed_time":70.14283379,"remaining_time":252.6002665,"test":[0.5231687739]}, -{"learn":[0.4425582941],"iteration":326,"passed_time":70.34791195,"remaining_time":252.3489319,"test":[0.5231483011]}, -{"learn":[0.4424612601],"iteration":327,"passed_time":70.5554264,"remaining_time":252.1065846,"test":[0.5231146588]}, -{"learn":[0.4422838046],"iteration":328,"passed_time":70.7706177,"remaining_time":251.891773,"test":[0.5230951039]}, -{"learn":[0.4422160868],"iteration":329,"passed_time":70.97467081,"remaining_time":251.6374692,"test":[0.5230877608]}, -{"learn":[0.4421166229],"iteration":330,"passed_time":71.18393847,"remaining_time":251.4018854,"test":[0.523061701]}, -{"learn":[0.441977886],"iteration":331,"passed_time":71.41116059,"remaining_time":251.2296252,"test":[0.5230415874]}, -{"learn":[0.4418363758],"iteration":332,"passed_time":71.62956294,"remaining_time":251.026126,"test":[0.5229807677]}, -{"learn":[0.4417645907],"iteration":333,"passed_time":71.83931094,"remaining_time":250.792325,"test":[0.5229519143]}, -{"learn":[0.441594768],"iteration":334,"passed_time":72.05350908,"remaining_time":250.5741435,"test":[0.5229005928]}, -{"learn":[0.4413897658],"iteration":335,"passed_time":72.27127429,"remaining_time":250.3683431,"test":[0.5227854185]}, -{"learn":[0.4412564167],"iteration":336,"passed_time":72.48459261,"remaining_time":250.1471252,"test":[0.5227791929]}, -{"learn":[0.4411897553],"iteration":337,"passed_time":72.69149127,"remaining_time":249.9038842,"test":[0.5227785145]}, -{"learn":[0.4409984605],"iteration":338,"passed_time":72.90594307,"remaining_time":249.6867254,"test":[0.5227113096]}, -{"learn":[0.4408126062],"iteration":339,"passed_time":73.12027871,"remaining_time":249.4691862,"test":[0.5227014922]}, -{"learn":[0.4407580675],"iteration":340,"passed_time":73.32371639,"remaining_time":249.2146255,"test":[0.5226839726]}, -{"learn":[0.440634781],"iteration":341,"passed_time":73.53363207,"remaining_time":248.9822981,"test":[0.5226773878]}, -{"learn":[0.4405851283],"iteration":342,"passed_time":73.73642021,"remaining_time":248.7260588,"test":[0.5226805805]}, -{"learn":[0.4404759981],"iteration":343,"passed_time":73.95494873,"remaining_time":248.5230254,"test":[0.5226501307]}, -{"learn":[0.4403500704],"iteration":344,"passed_time":74.17107234,"remaining_time":248.3118509,"test":[0.5226338483]}, -{"learn":[0.4402948978],"iteration":345,"passed_time":74.37558119,"remaining_time":248.0619095,"test":[0.5226316933]}, -{"learn":[0.4401106547],"iteration":346,"passed_time":74.58788369,"remaining_time":247.8381265,"test":[0.522656516]}, -{"learn":[0.4397904477],"iteration":347,"passed_time":74.80778869,"remaining_time":247.6395764,"test":[0.5226600678]}, -{"learn":[0.4397326341],"iteration":348,"passed_time":75.01209471,"remaining_time":247.3894585,"test":[0.5226367217]}, -{"learn":[0.4396824532],"iteration":349,"passed_time":75.21596581,"remaining_time":247.1381734,"test":[0.5225947386]}, -{"learn":[0.439624032],"iteration":350,"passed_time":75.42224509,"remaining_time":246.8950416,"test":[0.5226150118]}, -{"learn":[0.4394675733],"iteration":351,"passed_time":75.66408973,"remaining_time":246.7681108,"test":[0.5225920647]}, -{"learn":[0.4394047943],"iteration":352,"passed_time":75.87934364,"remaining_time":246.5541279,"test":[0.5225688783]}, -{"learn":[0.4393420418],"iteration":353,"passed_time":76.09058534,"remaining_time":246.3271492,"test":[0.5225403042]}, -{"learn":[0.4392639445],"iteration":354,"passed_time":76.29449611,"remaining_time":246.0766142,"test":[0.5225377102]}, -{"learn":[0.4391883562],"iteration":355,"passed_time":76.49915018,"remaining_time":245.8287298,"test":[0.5225459711]}, -{"learn":[0.4391398392],"iteration":356,"passed_time":76.70743543,"remaining_time":245.5927134,"test":[0.5225433372]}, -{"learn":[0.4389915678],"iteration":357,"passed_time":76.92544497,"remaining_time":245.3878719,"test":[0.5225567462]}, -{"learn":[0.4387959944],"iteration":358,"passed_time":77.13400271,"remaining_time":245.1529167,"test":[0.5225509995]}, -{"learn":[0.438557371],"iteration":359,"passed_time":77.34720003,"remaining_time":244.9328001,"test":[0.5225457317]}, -{"learn":[0.4385087747],"iteration":360,"passed_time":77.55081607,"remaining_time":244.6824917,"test":[0.5225362735]}, -{"learn":[0.4383729694],"iteration":361,"passed_time":77.75979666,"remaining_time":244.4493055,"test":[0.5225390271]}, -{"learn":[0.4383069419],"iteration":362,"passed_time":77.96371269,"remaining_time":244.2003893,"test":[0.5225502413]}, -{"learn":[0.4381990265],"iteration":363,"passed_time":78.17208503,"remaining_time":243.965628,"test":[0.5225557086]}, -{"learn":[0.43801238],"iteration":364,"passed_time":78.3862752,"remaining_time":243.7491023,"test":[0.5225313249]}, -{"learn":[0.4378701568],"iteration":365,"passed_time":78.59686483,"remaining_time":243.5214336,"test":[0.5224742965]}, -{"learn":[0.4378172555],"iteration":366,"passed_time":78.80121226,"remaining_time":243.2745872,"test":[0.5224713433]}, -{"learn":[0.4377448894],"iteration":367,"passed_time":79.0218156,"remaining_time":243.0779762,"test":[0.5224824776]}, -{"learn":[0.4374820735],"iteration":368,"passed_time":79.24130691,"remaining_time":242.8778269,"test":[0.5223904899]}, -{"learn":[0.437295876],"iteration":369,"passed_time":79.45799234,"remaining_time":242.6690036,"test":[0.5223451146]}, -{"learn":[0.4372092215],"iteration":370,"passed_time":79.6758244,"remaining_time":242.4636274,"test":[0.5222921568]}, -{"learn":[0.4370238163],"iteration":371,"passed_time":79.89964997,"remaining_time":242.276358,"test":[0.5222611883]}, -{"learn":[0.436981902],"iteration":372,"passed_time":80.12221775,"remaining_time":242.0850922,"test":[0.5222393587]}, -{"learn":[0.4368016998],"iteration":373,"passed_time":80.3363049,"remaining_time":241.8681265,"test":[0.52223014]}, -{"learn":[0.4366709917],"iteration":374,"passed_time":80.54500939,"remaining_time":241.6350282,"test":[0.5222536457]}, -{"learn":[0.4365565529],"iteration":375,"passed_time":80.75362562,"remaining_time":241.4017957,"test":[0.5223180172]}, -{"learn":[0.4363945478],"iteration":376,"passed_time":80.9662181,"remaining_time":241.1805383,"test":[0.5222929151]}, -{"learn":[0.4363409599],"iteration":377,"passed_time":81.17094127,"remaining_time":240.9359685,"test":[0.5222633433]}, -{"learn":[0.4362627569],"iteration":378,"passed_time":81.3760377,"remaining_time":240.6927131,"test":[0.5222620663]}, -{"learn":[0.4361928206],"iteration":379,"passed_time":81.57966136,"remaining_time":240.4453177,"test":[0.5223006971]}, -{"learn":[0.4361403683],"iteration":380,"passed_time":81.78539885,"remaining_time":240.2043604,"test":[0.5223303886]}, -{"learn":[0.4359903539],"iteration":381,"passed_time":82.00329977,"remaining_time":239.9991862,"test":[0.5222801845]}, -{"learn":[0.4357839519],"iteration":382,"passed_time":82.22043156,"remaining_time":239.7917025,"test":[0.5223078406]}, -{"learn":[0.4357237876],"iteration":383,"passed_time":82.42314338,"remaining_time":239.5422604,"test":[0.5223389688]}, -{"learn":[0.4356793379],"iteration":384,"passed_time":82.62773229,"remaining_time":239.2984974,"test":[0.5223291515]}, -{"learn":[0.4355541498],"iteration":385,"passed_time":82.83762727,"remaining_time":239.0702507,"test":[0.5223007769]}, -{"learn":[0.4353642019],"iteration":386,"passed_time":83.05126086,"remaining_time":238.852851,"test":[0.5222951898]}, -{"learn":[0.4352438734],"iteration":387,"passed_time":83.26045745,"remaining_time":238.6227543,"test":[0.5222391192]}, -{"learn":[0.4351800644],"iteration":388,"passed_time":83.46642758,"remaining_time":238.3835502,"test":[0.5223085989]}, -{"learn":[0.4350318459],"iteration":389,"passed_time":83.68036085,"remaining_time":238.1671809,"test":[0.5222770317]}, -{"learn":[0.4348217199],"iteration":390,"passed_time":83.89177884,"remaining_time":237.9436899,"test":[0.5223066434]}, -{"learn":[0.4347826317],"iteration":391,"passed_time":84.11260227,"remaining_time":237.7468452,"test":[0.5222492159]}, -{"learn":[0.4346708867],"iteration":392,"passed_time":84.32047257,"remaining_time":237.5133922,"test":[0.52228214]}, -{"learn":[0.434634598],"iteration":393,"passed_time":84.52383669,"remaining_time":237.2674198,"test":[0.5222827785]}, -{"learn":[0.4344987662],"iteration":394,"passed_time":84.76677866,"remaining_time":237.1323808,"test":[0.5222562397]}, -{"learn":[0.4343509967],"iteration":395,"passed_time":84.97668424,"remaining_time":236.9046955,"test":[0.5222660571]}, -{"learn":[0.4342753292],"iteration":396,"passed_time":85.18881281,"remaining_time":236.6832759,"test":[0.5222762336]}, -{"learn":[0.4341274276],"iteration":397,"passed_time":85.39937454,"remaining_time":236.4575647,"test":[0.5222696488]}, -{"learn":[0.4339128118],"iteration":398,"passed_time":85.6105665,"remaining_time":236.2336685,"test":[0.5221320862]}, -{"learn":[0.4338583523],"iteration":399,"passed_time":85.81337079,"remaining_time":235.9867697,"test":[0.5221256211]}, -{"learn":[0.4336283917],"iteration":400,"passed_time":86.03224374,"remaining_time":235.7841294,"test":[0.5221399082]}, -{"learn":[0.4335961703],"iteration":401,"passed_time":86.23464097,"remaining_time":235.5364074,"test":[0.522140866]}, -{"learn":[0.4334266645],"iteration":402,"passed_time":86.45127517,"remaining_time":235.3276647,"test":[0.5221340417]}, -{"learn":[0.4333842749],"iteration":403,"passed_time":86.65311476,"remaining_time":235.078747,"test":[0.5220711868]}, -{"learn":[0.433300552],"iteration":404,"passed_time":86.85605789,"remaining_time":234.8330454,"test":[0.5221006787]}, -{"learn":[0.4332627579],"iteration":405,"passed_time":87.0611953,"remaining_time":234.5934671,"test":[0.5221415444]}, -{"learn":[0.4332207644],"iteration":406,"passed_time":87.26593982,"remaining_time":234.353003,"test":[0.5221244239]}, -{"learn":[0.4331742546],"iteration":407,"passed_time":87.47074681,"remaining_time":234.1128812,"test":[0.5221450164]}, -{"learn":[0.4331175502],"iteration":408,"passed_time":87.67623363,"remaining_time":233.8747455,"test":[0.5221361169]}, -{"learn":[0.4330016059],"iteration":409,"passed_time":87.88625562,"remaining_time":233.6488259,"test":[0.5221247831]}, -{"learn":[0.4329391703],"iteration":410,"passed_time":88.1021335,"remaining_time":233.4384997,"test":[0.5221431407]}, -{"learn":[0.4327436761],"iteration":411,"passed_time":88.33286598,"remaining_time":233.2673742,"test":[0.5221476104]}, -{"learn":[0.4326503132],"iteration":412,"passed_time":88.54745586,"remaining_time":233.0534734,"test":[0.522193704]}, -{"learn":[0.4326064445],"iteration":413,"passed_time":88.75546066,"remaining_time":232.8222953,"test":[0.5221898729]}, -{"learn":[0.4325310147],"iteration":414,"passed_time":88.95929179,"remaining_time":232.5803171,"test":[0.5222235951]}, -{"learn":[0.4322932893],"iteration":415,"passed_time":89.18501335,"remaining_time":232.3955636,"test":[0.5222830578]}, -{"learn":[0.4321530205],"iteration":416,"passed_time":89.39244899,"remaining_time":232.1631229,"test":[0.5222495751]}, -{"learn":[0.4320995646],"iteration":417,"passed_time":89.5957362,"remaining_time":231.9200636,"test":[0.5222620663]}, -{"learn":[0.4320641475],"iteration":418,"passed_time":89.79808474,"remaining_time":231.6747723,"test":[0.5222605099]}, -{"learn":[0.4318961471],"iteration":419,"passed_time":90.0087367,"remaining_time":231.4510372,"test":[0.5222401967]}, -{"learn":[0.431762032],"iteration":420,"passed_time":90.22152765,"remaining_time":231.2328464,"test":[0.5223113924]}, -{"learn":[0.4316429712],"iteration":421,"passed_time":90.429538,"remaining_time":231.0024691,"test":[0.5222913587]}, -{"learn":[0.4316029322],"iteration":422,"passed_time":90.63332669,"remaining_time":230.7614488,"test":[0.5223000985]}, -{"learn":[0.4314649611],"iteration":423,"passed_time":90.84374012,"remaining_time":230.537416,"test":[0.5222907202]}, -{"learn":[0.4313357849],"iteration":424,"passed_time":91.05443958,"remaining_time":230.3141707,"test":[0.5222583947]}, -{"learn":[0.4312889318],"iteration":425,"passed_time":91.26381978,"remaining_time":230.0876583,"test":[0.5222463426]}, -{"learn":[0.4311176301],"iteration":426,"passed_time":91.48438264,"remaining_time":229.8893269,"test":[0.5221930256]}, -{"learn":[0.4310804434],"iteration":427,"passed_time":91.68641059,"remaining_time":229.6444676,"test":[0.5221907908]}, -{"learn":[0.4309550439],"iteration":428,"passed_time":91.89042499,"remaining_time":229.4047673,"test":[0.5221801753]}, -{"learn":[0.4308267921],"iteration":429,"passed_time":92.09898827,"remaining_time":229.1765522,"test":[0.5222163318]}, -{"learn":[0.4306758797],"iteration":430,"passed_time":92.32109712,"remaining_time":228.9820251,"test":[0.5222692497]}, -{"learn":[0.430564214],"iteration":431,"passed_time":92.55081837,"remaining_time":228.8061899,"test":[0.5222568383]}, -{"learn":[0.4305215338],"iteration":432,"passed_time":92.75444592,"remaining_time":228.5658055,"test":[0.52226478]}, -{"learn":[0.4304761861],"iteration":433,"passed_time":92.95713727,"remaining_time":228.3232911,"test":[0.5222783088]}, -{"learn":[0.4303507603],"iteration":434,"passed_time":93.1708367,"remaining_time":228.1079105,"test":[0.5222840954]}, -{"learn":[0.4302473612],"iteration":435,"passed_time":93.38517138,"remaining_time":227.8940879,"test":[0.5222134984]}, -{"learn":[0.4302070316],"iteration":436,"passed_time":93.58930039,"remaining_time":227.6554378,"test":[0.52221713]}, -{"learn":[0.4301724332],"iteration":437,"passed_time":93.79648265,"remaining_time":227.4243483,"test":[0.5222277056]}, -{"learn":[0.4300275953],"iteration":438,"passed_time":94.0038674,"remaining_time":227.1938572,"test":[0.5222634631]}, -{"learn":[0.4299943702],"iteration":439,"passed_time":94.23371294,"remaining_time":227.0175812,"test":[0.5223040494]}, -{"learn":[0.4298901524],"iteration":440,"passed_time":94.44963657,"remaining_time":226.8076307,"test":[0.5222941522]}, -{"learn":[0.4298619719],"iteration":441,"passed_time":94.65202482,"remaining_time":226.565254,"test":[0.5222502934]}, -{"learn":[0.4297966311],"iteration":442,"passed_time":94.86756504,"remaining_time":226.3544385,"test":[0.5222784285]}, -{"learn":[0.4295941908],"iteration":443,"passed_time":95.09021783,"remaining_time":226.1605181,"test":[0.5222565989]}, -{"learn":[0.4293884755],"iteration":444,"passed_time":95.29992534,"remaining_time":225.9357781,"test":[0.5222131392]}, -{"learn":[0.4292060811],"iteration":445,"passed_time":95.510937,"remaining_time":225.7141874,"test":[0.5222200433]}, -{"learn":[0.4291471318],"iteration":446,"passed_time":95.71507987,"remaining_time":225.4764633,"test":[0.5222137777]}, -{"learn":[0.4290125149],"iteration":447,"passed_time":95.91936096,"remaining_time":225.2392137,"test":[0.5221952205]}, -{"learn":[0.4288033134],"iteration":448,"passed_time":96.14019167,"remaining_time":225.0408495,"test":[0.5222177685]}, -{"learn":[0.4287767175],"iteration":449,"passed_time":96.35095311,"remaining_time":224.8188906,"test":[0.5222333326]}, -{"learn":[0.4287522081],"iteration":450,"passed_time":96.5597974,"remaining_time":224.5925221,"test":[0.5222451054]}, -{"learn":[0.4287100297],"iteration":451,"passed_time":96.77169136,"remaining_time":224.3733021,"test":[0.5222368046]}, -{"learn":[0.4284893658],"iteration":452,"passed_time":96.98769612,"remaining_time":224.1636155,"test":[0.5221670056]}, -{"learn":[0.4284410865],"iteration":453,"passed_time":97.1954738,"remaining_time":223.9349462,"test":[0.5221495659]}, -{"learn":[0.4284061448],"iteration":454,"passed_time":97.40168562,"remaining_time":223.7027725,"test":[0.5221062658]}, -{"learn":[0.428283677],"iteration":455,"passed_time":97.60761641,"remaining_time":223.4700691,"test":[0.5221362366]}, -{"learn":[0.42796347],"iteration":456,"passed_time":97.83021101,"remaining_time":223.2755144,"test":[0.5221121722]}, -{"learn":[0.4279105952],"iteration":457,"passed_time":98.03536187,"remaining_time":223.0411508,"test":[0.5220806848]}, -{"learn":[0.4276986997],"iteration":458,"passed_time":98.25440598,"remaining_time":222.838424,"test":[0.5220582167]}, -{"learn":[0.4275380152],"iteration":459,"passed_time":98.46506595,"remaining_time":222.6166708,"test":[0.5221086603]}, -{"learn":[0.4274078618],"iteration":460,"passed_time":98.67278728,"remaining_time":222.3883427,"test":[0.5220843165]}, -{"learn":[0.4272547308],"iteration":461,"passed_time":98.88326351,"remaining_time":222.1662933,"test":[0.5220927769]}, -{"learn":[0.4272262598],"iteration":462,"passed_time":99.0837289,"remaining_time":221.9218723,"test":[0.5220818422]}, -{"learn":[0.4271944609],"iteration":463,"passed_time":99.28833123,"remaining_time":221.6868775,"test":[0.5220434907]}, -{"learn":[0.4271767391],"iteration":464,"passed_time":99.33595715,"remaining_time":221.1026143,"test":[0.522045526]}, -{"learn":[0.4271252641],"iteration":465,"passed_time":99.54019114,"remaining_time":220.8681494,"test":[0.5220438099]}, -{"learn":[0.4269907529],"iteration":466,"passed_time":99.74994554,"remaining_time":220.6460251,"test":[0.5220508337]}, -{"learn":[0.4269561545],"iteration":467,"passed_time":99.95657062,"remaining_time":220.4170532,"test":[0.5220992819]}, -{"learn":[0.4267656783],"iteration":468,"passed_time":100.1935219,"remaining_time":220.2548423,"test":[0.5220674753]}, -{"learn":[0.4266574461],"iteration":469,"passed_time":100.418076,"remaining_time":220.0651453,"test":[0.5220163533]}, -{"learn":[0.4266171429],"iteration":470,"passed_time":100.6502161,"remaining_time":219.8918734,"test":[0.5219874999]}, -{"learn":[0.4264591523],"iteration":471,"passed_time":100.8810043,"remaining_time":219.7154076,"test":[0.5220094492]}, -{"learn":[0.426324509],"iteration":472,"passed_time":101.0916121,"remaining_time":219.4948955,"test":[0.5219871806]}, -{"learn":[0.4262026222],"iteration":473,"passed_time":101.3004473,"remaining_time":219.2705884,"test":[0.5219403287]}, -{"learn":[0.4261693444],"iteration":474,"passed_time":101.5050295,"remaining_time":219.037169,"test":[0.521902935]}, -{"learn":[0.4259847579],"iteration":475,"passed_time":101.7234315,"remaining_time":218.8336006,"test":[0.5218628675]}, -{"learn":[0.4259156403],"iteration":476,"passed_time":101.9253522,"remaining_time":218.5946234,"test":[0.5218597946]}, -{"learn":[0.4258481602],"iteration":477,"passed_time":102.1293143,"remaining_time":218.3601658,"test":[0.5218983456]}, -{"learn":[0.4258218548],"iteration":478,"passed_time":102.3319364,"remaining_time":218.1229792,"test":[0.5219061277]}, -{"learn":[0.4257969493],"iteration":479,"passed_time":102.5347374,"remaining_time":217.886317,"test":[0.5218939558]}, -{"learn":[0.4257487756],"iteration":480,"passed_time":102.7418339,"remaining_time":217.6588955,"test":[0.5219066864]}, -{"learn":[0.4256836461],"iteration":481,"passed_time":102.9483646,"remaining_time":217.4303635,"test":[0.5219238867]}, -{"learn":[0.4256607742],"iteration":482,"passed_time":103.1606265,"remaining_time":217.21399,"test":[0.521889965]}, -{"learn":[0.4255501649],"iteration":483,"passed_time":103.3739904,"remaining_time":216.9999468,"test":[0.521844869]}, -{"learn":[0.4255061642],"iteration":484,"passed_time":103.5782457,"remaining_time":216.766844,"test":[0.5218519726]}, -{"learn":[0.4253049916],"iteration":485,"passed_time":103.7932079,"remaining_time":216.5561992,"test":[0.5218498575]}, -{"learn":[0.4251170245],"iteration":486,"passed_time":104.002997,"remaining_time":216.3347761,"test":[0.5218545267]}, -{"learn":[0.4249721602],"iteration":487,"passed_time":104.2163309,"remaining_time":216.1207518,"test":[0.5217666894]}, -{"learn":[0.4249461189],"iteration":488,"passed_time":104.4191904,"remaining_time":215.8850746,"test":[0.5217849274]}, -{"learn":[0.4249260201],"iteration":489,"passed_time":104.6301125,"remaining_time":215.6661502,"test":[0.5217765467]}, -{"learn":[0.424814645],"iteration":490,"passed_time":104.8473614,"remaining_time":215.4602599,"test":[0.5217925099]}, -{"learn":[0.4247909543],"iteration":491,"passed_time":105.0673067,"remaining_time":215.259848,"test":[0.5218065974]}, -{"learn":[0.4247375248],"iteration":492,"passed_time":105.273126,"remaining_time":215.0305027,"test":[0.5217701215]}, -{"learn":[0.4246067376],"iteration":493,"passed_time":105.482074,"remaining_time":214.8076245,"test":[0.5218237178]}, -{"learn":[0.4245252596],"iteration":494,"passed_time":105.692536,"remaining_time":214.587876,"test":[0.5218390824]}, -{"learn":[0.4243307162],"iteration":495,"passed_time":105.9059334,"remaining_time":214.3741071,"test":[0.5217997731]}, -{"learn":[0.4242090143],"iteration":496,"passed_time":106.1186945,"remaining_time":214.1590555,"test":[0.5218265513]}, -{"learn":[0.4241103428],"iteration":497,"passed_time":106.3329792,"remaining_time":213.9470786,"test":[0.5218737624]}, -{"learn":[0.4240749785],"iteration":498,"passed_time":106.5340114,"remaining_time":213.7085077,"test":[0.5218910425]}, -{"learn":[0.423876315],"iteration":499,"passed_time":106.7472701,"remaining_time":213.4945402,"test":[0.52187911]}, -{"learn":[0.4236667437],"iteration":500,"passed_time":106.9613022,"remaining_time":213.2821176,"test":[0.5218104684]}, -{"learn":[0.4235874314],"iteration":501,"passed_time":107.1683723,"remaining_time":213.0558478,"test":[0.5218612712]}, -{"learn":[0.4235353754],"iteration":502,"passed_time":107.3749381,"remaining_time":212.8286546,"test":[0.521854846]}, -{"learn":[0.4234143602],"iteration":503,"passed_time":107.5844873,"remaining_time":212.6074393,"test":[0.5218411177]}, -{"learn":[0.4233028793],"iteration":504,"passed_time":107.7910702,"remaining_time":212.3804253,"test":[0.5218925989]}, -{"learn":[0.423176714],"iteration":505,"passed_time":107.9939197,"remaining_time":212.1461584,"test":[0.5218558836]}, -{"learn":[0.4231508576],"iteration":506,"passed_time":108.2074939,"remaining_time":211.9330205,"test":[0.5218650624]}, -{"learn":[0.4229899882],"iteration":507,"passed_time":108.4167864,"remaining_time":211.7115199,"test":[0.5218238775]}, -{"learn":[0.4228718782],"iteration":508,"passed_time":108.6394747,"remaining_time":211.5161481,"test":[0.5218550455]}, -{"learn":[0.4228359329],"iteration":509,"passed_time":108.8545844,"remaining_time":211.305958,"test":[0.521825394]}, -{"learn":[0.4227991952],"iteration":510,"passed_time":109.0642878,"remaining_time":211.0852849,"test":[0.5218181706]}, -{"learn":[0.4227724144],"iteration":511,"passed_time":109.28476,"remaining_time":210.8854352,"test":[0.5218546066]}, -{"learn":[0.4225436952],"iteration":512,"passed_time":109.5214753,"remaining_time":210.7167567,"test":[0.5218829412]}, -{"learn":[0.4224996416],"iteration":513,"passed_time":109.7290791,"remaining_time":210.4919689,"test":[0.5218839788]}, -{"learn":[0.4223835653],"iteration":514,"passed_time":109.9470032,"remaining_time":210.2869867,"test":[0.5218613909]}, -{"learn":[0.4223577353],"iteration":515,"passed_time":110.1575744,"remaining_time":210.0679326,"test":[0.5218823825]}, -{"learn":[0.4222493446],"iteration":516,"passed_time":110.3645408,"remaining_time":209.8420573,"test":[0.5218802673]}, -{"learn":[0.4222281101],"iteration":517,"passed_time":110.5688433,"remaining_time":209.6112049,"test":[0.5219119143]}, -{"learn":[0.422172198],"iteration":518,"passed_time":110.7701722,"remaining_time":209.3748341,"test":[0.5218955521]}, -{"learn":[0.4221387881],"iteration":519,"passed_time":110.9724019,"remaining_time":209.1402959,"test":[0.5219027355]}, -{"learn":[0.4221057744],"iteration":520,"passed_time":111.1812834,"remaining_time":208.9183809,"test":[0.5219047309]}, -{"learn":[0.4219720291],"iteration":521,"passed_time":111.3860637,"remaining_time":208.688832,"test":[0.5218664193]}, -{"learn":[0.4218148572],"iteration":522,"passed_time":111.5987443,"remaining_time":208.4741361,"test":[0.5218799481]}, -{"learn":[0.421678867],"iteration":523,"passed_time":111.8119444,"remaining_time":208.2604155,"test":[0.5218879297]}, -{"learn":[0.4215360891],"iteration":524,"passed_time":112.0283365,"remaining_time":208.0526249,"test":[0.5218919604]}, -{"learn":[0.4214808109],"iteration":525,"passed_time":112.2347162,"remaining_time":207.8262616,"test":[0.5219174216]}, -{"learn":[0.421386418],"iteration":526,"passed_time":112.4686428,"remaining_time":207.6508338,"test":[0.521921572]}, -{"learn":[0.4212872182],"iteration":527,"passed_time":112.6744949,"remaining_time":207.423502,"test":[0.521982232]}, -{"learn":[0.4212450663],"iteration":528,"passed_time":112.8770865,"remaining_time":207.1902665,"test":[0.5220316779]}, -{"learn":[0.421218312],"iteration":529,"passed_time":113.1033054,"remaining_time":207.0003891,"test":[0.5220599726]}, -{"learn":[0.421077647],"iteration":530,"passed_time":113.3161359,"remaining_time":206.7859429,"test":[0.5220624868]}, -{"learn":[0.4209894342],"iteration":531,"passed_time":113.5325177,"remaining_time":206.5779645,"test":[0.5220547447]}, -{"learn":[0.4209177812],"iteration":532,"passed_time":113.7416202,"remaining_time":206.3567481,"test":[0.5220632451]}, -{"learn":[0.4208865898],"iteration":533,"passed_time":113.9464601,"remaining_time":206.1278661,"test":[0.5220848752]}, -{"learn":[0.4208590167],"iteration":534,"passed_time":114.1474219,"remaining_time":205.8920787,"test":[0.5220681937]}, -{"learn":[0.4207220229],"iteration":535,"passed_time":114.3558661,"remaining_time":205.6698786,"test":[0.5220366664]}, -{"learn":[0.4205735666],"iteration":536,"passed_time":114.5725608,"remaining_time":205.4625253,"test":[0.5220554231]}, -{"learn":[0.4205432204],"iteration":537,"passed_time":114.7737867,"remaining_time":205.2274773,"test":[0.5220509135]}, -{"learn":[0.4205065883],"iteration":538,"passed_time":114.9823051,"remaining_time":205.005557,"test":[0.5220928568]}, -{"learn":[0.4204169494],"iteration":539,"passed_time":115.1927305,"remaining_time":204.7870765,"test":[0.5221466526]}, -{"learn":[0.4203891914],"iteration":540,"passed_time":115.3997273,"remaining_time":204.562548,"test":[0.5221287739]}, -{"learn":[0.4202709494],"iteration":541,"passed_time":115.6195589,"remaining_time":204.3607702,"test":[0.5221223487]}, -{"learn":[0.420224017],"iteration":542,"passed_time":115.8249079,"remaining_time":204.1334012,"test":[0.5221388706]}, -{"learn":[0.4201979758],"iteration":543,"passed_time":116.0273892,"remaining_time":203.9010737,"test":[0.5221799358]}, -{"learn":[0.4201635359],"iteration":544,"passed_time":116.229706,"remaining_time":203.6685674,"test":[0.5221457746]}, -{"learn":[0.4201325029],"iteration":545,"passed_time":116.4327778,"remaining_time":203.4374909,"test":[0.5221054677]}, -{"learn":[0.4201024736],"iteration":546,"passed_time":116.636416,"remaining_time":203.2075035,"test":[0.5221112144]}, -{"learn":[0.4199849975],"iteration":547,"passed_time":116.847819,"remaining_time":202.9911017,"test":[0.5220737808]}, -{"learn":[0.4198871712],"iteration":548,"passed_time":117.0579428,"remaining_time":202.772502,"test":[0.5220705881]}, -{"learn":[0.4197798368],"iteration":549,"passed_time":117.3019895,"remaining_time":202.6125273,"test":[0.5220880678]}, -{"learn":[0.4196252004],"iteration":550,"passed_time":117.5235767,"remaining_time":202.413565,"test":[0.5220916196]}, -{"learn":[0.4195158061],"iteration":551,"passed_time":117.7338517,"remaining_time":202.1950931,"test":[0.5221136089]}, -{"learn":[0.4193934703],"iteration":552,"passed_time":117.9480233,"remaining_time":201.9833237,"test":[0.5220915797]}, -{"learn":[0.4192237533],"iteration":553,"passed_time":118.1667645,"remaining_time":201.7793487,"test":[0.5220772129]}, -{"learn":[0.4191110047],"iteration":554,"passed_time":118.3926687,"remaining_time":201.587517,"test":[0.5221100172]}, -{"learn":[0.419012386],"iteration":555,"passed_time":118.6074088,"remaining_time":201.3766078,"test":[0.5220727033]}, -{"learn":[0.4189145333],"iteration":556,"passed_time":118.8113215,"remaining_time":201.1473541,"test":[0.5220840371]}, -{"learn":[0.4188031845],"iteration":557,"passed_time":119.0142518,"remaining_time":200.9165327,"test":[0.5220930164]}, -{"learn":[0.4185742011],"iteration":558,"passed_time":119.2357886,"remaining_time":200.7171325,"test":[0.5220205436]}, -{"learn":[0.4185236769],"iteration":559,"passed_time":119.4401694,"remaining_time":200.4888557,"test":[0.5220209826]}, -{"learn":[0.4184201458],"iteration":560,"passed_time":119.6482905,"remaining_time":200.2669247,"test":[0.522004341]}, -{"learn":[0.4183944214],"iteration":561,"passed_time":119.8526739,"remaining_time":200.0388045,"test":[0.5220604515]}, -{"learn":[0.4183447688],"iteration":562,"passed_time":120.0543716,"remaining_time":199.8062988,"test":[0.5220947723]}, -{"learn":[0.4182207955],"iteration":563,"passed_time":120.2661929,"remaining_time":199.5907032,"test":[0.5220045007]}, -{"learn":[0.4180463509],"iteration":564,"passed_time":120.4836934,"remaining_time":199.3845191,"test":[0.5220480801]}, -{"learn":[0.41787703],"iteration":565,"passed_time":120.6965569,"remaining_time":199.1706434,"test":[0.522024734]}, -{"learn":[0.4177670546],"iteration":566,"passed_time":120.9066805,"remaining_time":198.9522626,"test":[0.5220192267]}, -{"learn":[0.4177384779],"iteration":567,"passed_time":121.1116605,"remaining_time":198.7254711,"test":[0.5220286848]}, -{"learn":[0.4175663046],"iteration":568,"passed_time":121.3445522,"remaining_time":198.5444254,"test":[0.5220263702]}, -{"learn":[0.4174468476],"iteration":569,"passed_time":121.5686459,"remaining_time":198.3488434,"test":[0.5220041016]}, -{"learn":[0.4172682036],"iteration":570,"passed_time":121.8164141,"remaining_time":198.1916789,"test":[0.5220519112]}, -{"learn":[0.4171848769],"iteration":571,"passed_time":122.0230116,"remaining_time":197.9674034,"test":[0.5221098974]}, -{"learn":[0.417080395],"iteration":572,"passed_time":122.2303868,"remaining_time":197.7444477,"test":[0.5221240647]}, -{"learn":[0.4170362886],"iteration":573,"passed_time":122.4282092,"remaining_time":197.5061354,"test":[0.5221256611]}, -{"learn":[0.4169296674],"iteration":574,"passed_time":122.6417976,"remaining_time":197.2933266,"test":[0.5221025145]}, -{"learn":[0.4168813353],"iteration":575,"passed_time":122.8451356,"remaining_time":197.0640716,"test":[0.5221338821]}, -{"learn":[0.4168014156],"iteration":576,"passed_time":123.0525376,"remaining_time":196.8414077,"test":[0.5221315674]}, -{"learn":[0.4166510314],"iteration":577,"passed_time":123.265679,"remaining_time":196.6279516,"test":[0.5221554324]}, -{"learn":[0.4165654861],"iteration":578,"passed_time":123.4706621,"remaining_time":196.4015196,"test":[0.5222276258]}, -{"learn":[0.4164509152],"iteration":579,"passed_time":123.6782914,"remaining_time":196.1793587,"test":[0.5222340908]}, -{"learn":[0.4163564959],"iteration":580,"passed_time":123.8844284,"remaining_time":195.9548875,"test":[0.5222625452]}, -{"learn":[0.4163250932],"iteration":581,"passed_time":124.0892174,"remaining_time":195.7283532,"test":[0.5222746772]}, -{"learn":[0.4161199854],"iteration":582,"passed_time":124.3119605,"remaining_time":195.5301334,"test":[0.522315503]}, -{"learn":[0.4160572857],"iteration":583,"passed_time":124.5317846,"remaining_time":195.3272512,"test":[0.5223301492]}, -{"learn":[0.4160180917],"iteration":584,"passed_time":124.7349994,"remaining_time":195.0983325,"test":[0.5223704561]}, -{"learn":[0.4159334181],"iteration":585,"passed_time":124.9399464,"remaining_time":194.872203,"test":[0.522333262]}, -{"learn":[0.4157927531],"iteration":586,"passed_time":125.1522334,"remaining_time":194.6575624,"test":[0.5223089182]}, -{"learn":[0.4156931836],"iteration":587,"passed_time":125.3607684,"remaining_time":194.4371102,"test":[0.5222736396]}, -{"learn":[0.4154709615],"iteration":588,"passed_time":125.5962772,"remaining_time":194.2584185,"test":[0.5223175383]}, -{"learn":[0.415443811],"iteration":589,"passed_time":125.823019,"remaining_time":194.0660124,"test":[0.5223279542]}, -{"learn":[0.4153113862],"iteration":590,"passed_time":126.0276476,"remaining_time":193.8394783,"test":[0.5223895321]}, -{"learn":[0.4152067722],"iteration":591,"passed_time":126.2372792,"remaining_time":193.6206917,"test":[0.5223692589]}, -{"learn":[0.4151275921],"iteration":592,"passed_time":126.4417727,"remaining_time":193.3940773,"test":[0.5223694984]}, -{"learn":[0.4149858178],"iteration":593,"passed_time":126.6500259,"remaining_time":193.1732719,"test":[0.5224234139]}, -{"learn":[0.4149130555],"iteration":594,"passed_time":126.8615764,"remaining_time":192.9575238,"test":[0.5224167493]}, -{"learn":[0.4146729531],"iteration":595,"passed_time":127.0853141,"remaining_time":192.7602751,"test":[0.522434229]}, -{"learn":[0.4144590504],"iteration":596,"passed_time":127.3029649,"remaining_time":192.5537308,"test":[0.5223872973]}, -{"learn":[0.4143572096],"iteration":597,"passed_time":127.5491045,"remaining_time":192.3901209,"test":[0.5224070916]}, -{"learn":[0.4143302704],"iteration":598,"passed_time":127.7491871,"remaining_time":192.1569575,"test":[0.522377879]}, -{"learn":[0.4142094401],"iteration":599,"passed_time":127.9576759,"remaining_time":191.9365139,"test":[0.5223658268]}, -{"learn":[0.4140924129],"iteration":600,"passed_time":128.1637658,"remaining_time":191.7125216,"test":[0.5223325037]}, -{"learn":[0.4139592751],"iteration":601,"passed_time":128.3722047,"remaining_time":191.4920928,"test":[0.5223590026]}, -{"learn":[0.4138737827],"iteration":602,"passed_time":128.5784023,"remaining_time":191.2683696,"test":[0.522378318]}, -{"learn":[0.4138459455],"iteration":603,"passed_time":128.7837067,"remaining_time":191.0433795,"test":[0.5223461123]}, -{"learn":[0.4136362686],"iteration":604,"passed_time":129.0077605,"remaining_time":190.8461912,"test":[0.5223434784]}, -{"learn":[0.4136080088],"iteration":605,"passed_time":129.2101526,"remaining_time":190.6169578,"test":[0.5223463917]}, -{"learn":[0.4135243916],"iteration":606,"passed_time":129.4151721,"remaining_time":190.3916783,"test":[0.5223482274]}, -{"learn":[0.4133501054],"iteration":607,"passed_time":129.6246748,"remaining_time":190.1730426,"test":[0.5223358959]}, -{"learn":[0.4132487135],"iteration":608,"passed_time":129.8907793,"remaining_time":190.0372486,"test":[0.5223317854]}, -{"learn":[0.4131160511],"iteration":609,"passed_time":130.1383303,"remaining_time":189.8739574,"test":[0.5223568476]}, -{"learn":[0.4130051249],"iteration":610,"passed_time":130.380929,"remaining_time":189.7031847,"test":[0.5223685805]}, -{"learn":[0.4129740391],"iteration":611,"passed_time":130.6005907,"remaining_time":189.4988963,"test":[0.5223942812]}, -{"learn":[0.4128680782],"iteration":612,"passed_time":130.8083957,"remaining_time":189.2774012,"test":[0.5223829473]}, -{"learn":[0.412837415],"iteration":613,"passed_time":131.0093857,"remaining_time":189.0461169,"test":[0.5224129182]}, -{"learn":[0.4127119892],"iteration":614,"passed_time":131.2143968,"remaining_time":188.8207174,"test":[0.5223948]}, -{"learn":[0.4126791075],"iteration":615,"passed_time":131.4163495,"remaining_time":188.590995,"test":[0.5224285221]}, -{"learn":[0.4125751802],"iteration":616,"passed_time":131.6329015,"remaining_time":188.3822561,"test":[0.522424691]}, -{"learn":[0.4124947851],"iteration":617,"passed_time":131.8399089,"remaining_time":188.15987,"test":[0.52240777]}, -{"learn":[0.4123330706],"iteration":618,"passed_time":132.0549863,"remaining_time":187.9490192,"test":[0.5224196227]}, -{"learn":[0.4123064483],"iteration":619,"passed_time":132.2530748,"remaining_time":187.7140416,"test":[0.5223986311]}, -{"learn":[0.4121374971],"iteration":620,"passed_time":132.4636872,"remaining_time":187.4969099,"test":[0.5224034201]}, -{"learn":[0.4120678249],"iteration":621,"passed_time":132.6701686,"remaining_time":187.273968,"test":[0.5223694584]}, -{"learn":[0.4119360076],"iteration":622,"passed_time":132.8863661,"remaining_time":187.0647561,"test":[0.5223874569]}, -{"learn":[0.4118715384],"iteration":623,"passed_time":133.1017828,"remaining_time":186.8544258,"test":[0.5223998284]}, -{"learn":[0.4117428904],"iteration":624,"passed_time":133.345456,"remaining_time":186.6836384,"test":[0.5223842244]}, -{"learn":[0.4116109146],"iteration":625,"passed_time":133.5735411,"remaining_time":186.4908544,"test":[0.5223773203]}, -{"learn":[0.4115017844],"iteration":626,"passed_time":133.7875189,"remaining_time":186.2783158,"test":[0.5223327831]}, -{"learn":[0.411481131],"iteration":627,"passed_time":134.0320529,"remaining_time":186.1082009,"test":[0.5223507416]}, -{"learn":[0.4113286075],"iteration":628,"passed_time":134.259815,"remaining_time":185.9146246,"test":[0.5223174584]}, -{"learn":[0.4112463108],"iteration":629,"passed_time":134.4791466,"remaining_time":185.7092977,"test":[0.5223235244]}, -{"learn":[0.4110173538],"iteration":630,"passed_time":134.6993694,"remaining_time":185.5051538,"test":[0.5222586342]}, -{"learn":[0.4108394229],"iteration":631,"passed_time":134.9136933,"remaining_time":185.2928573,"test":[0.5222450655]}, -{"learn":[0.4105907634],"iteration":632,"passed_time":135.1327197,"remaining_time":185.0869952,"test":[0.5222296211]}, -{"learn":[0.4105139073],"iteration":633,"passed_time":135.3308854,"remaining_time":184.8525974,"test":[0.5222177685]}, -{"learn":[0.4101629052],"iteration":634,"passed_time":135.5548357,"remaining_time":184.6534376,"test":[0.5222282244]}, -{"learn":[0.4099986023],"iteration":635,"passed_time":135.7676373,"remaining_time":184.4390545,"test":[0.5222375628]}, -{"learn":[0.4098852727],"iteration":636,"passed_time":135.9874716,"remaining_time":184.2342041,"test":[0.5222393188]}, -{"learn":[0.4097632803],"iteration":637,"passed_time":136.2035831,"remaining_time":184.0242769,"test":[0.5222374032]}, -{"learn":[0.4096657973],"iteration":638,"passed_time":136.4274339,"remaining_time":183.8247583,"test":[0.5222169703]}, -{"learn":[0.4095620549],"iteration":639,"passed_time":136.6503039,"remaining_time":183.6238459,"test":[0.5222457838]}, -{"learn":[0.409416187],"iteration":640,"passed_time":136.8604854,"remaining_time":183.4058611,"test":[0.5221768629]}, -{"learn":[0.4093790267],"iteration":641,"passed_time":137.0676429,"remaining_time":183.1838592,"test":[0.5221749473]}, -{"learn":[0.4092723527],"iteration":642,"passed_time":137.2835737,"remaining_time":182.9735966,"test":[0.5221539957]}, -{"learn":[0.4091630639],"iteration":643,"passed_time":137.5339751,"remaining_time":182.8091346,"test":[0.5221308092]}, -{"learn":[0.4090574199],"iteration":644,"passed_time":137.7443969,"remaining_time":182.5914098,"test":[0.5220681139]}, -{"learn":[0.4089596992],"iteration":645,"passed_time":137.951932,"remaining_time":182.3698915,"test":[0.5220853541]}, -{"learn":[0.4088231016],"iteration":646,"passed_time":138.2272476,"remaining_time":182.2377777,"test":[0.5221375536]}, -{"learn":[0.4088041913],"iteration":647,"passed_time":138.4764773,"remaining_time":182.0709239,"test":[0.5221546741]}, -{"learn":[0.4086547843],"iteration":648,"passed_time":138.7071683,"remaining_time":181.8795073,"test":[0.5221431008]}, -{"learn":[0.4084877347],"iteration":649,"passed_time":138.9137598,"remaining_time":181.6564551,"test":[0.5221245436]}, -{"learn":[0.4084118559],"iteration":650,"passed_time":139.125069,"remaining_time":181.4396061,"test":[0.5221498852]}, -{"learn":[0.4082863508],"iteration":651,"passed_time":139.3525396,"remaining_time":181.2437939,"test":[0.5221640126]}, -{"learn":[0.4082518316],"iteration":652,"passed_time":139.5735289,"remaining_time":181.0394777,"test":[0.5221665667]}, -{"learn":[0.4080320657],"iteration":653,"passed_time":139.8499072,"remaining_time":180.9067606,"test":[0.5221492865]}, -{"learn":[0.4078376807],"iteration":654,"passed_time":140.06257,"remaining_time":180.691407,"test":[0.5221784991]}, -{"learn":[0.4076548902],"iteration":655,"passed_time":140.2845456,"remaining_time":180.4880434,"test":[0.5222281046]}, -{"learn":[0.407535354],"iteration":656,"passed_time":140.4986467,"remaining_time":180.2745192,"test":[0.522238361]}, -{"learn":[0.4074433909],"iteration":657,"passed_time":140.7095101,"remaining_time":180.0568504,"test":[0.5222958284]}, -{"learn":[0.4072420335],"iteration":658,"passed_time":140.9230402,"remaining_time":179.8426052,"test":[0.5223166204]}, -{"learn":[0.4071415924],"iteration":659,"passed_time":141.1291726,"remaining_time":179.6189469,"test":[0.5223578053]}, -{"learn":[0.407027312],"iteration":660,"passed_time":141.3330653,"remaining_time":179.392499,"test":[0.5223855812]}, -{"learn":[0.4069112621],"iteration":661,"passed_time":141.541929,"remaining_time":179.1724117,"test":[0.5223741676]}, -{"learn":[0.406769752],"iteration":662,"passed_time":141.7542664,"remaining_time":178.9567436,"test":[0.5223762827]}, -{"learn":[0.4065997444],"iteration":663,"passed_time":141.9643081,"remaining_time":178.7381952,"test":[0.5223555705]}, -{"learn":[0.40648932],"iteration":664,"passed_time":142.1758804,"remaining_time":178.5215942,"test":[0.5223429596]}, -{"learn":[0.4062134835],"iteration":665,"passed_time":142.4511564,"remaining_time":178.3847815,"test":[0.5222660172]}, -{"learn":[0.4061617443],"iteration":666,"passed_time":142.6826811,"remaining_time":178.1929135,"test":[0.5222825789]}, -{"learn":[0.4060348659],"iteration":667,"passed_time":142.8918509,"remaining_time":177.9730838,"test":[0.5222708061]}, -{"learn":[0.4060129712],"iteration":668,"passed_time":143.0936843,"remaining_time":177.7441729,"test":[0.5222718038]}, -{"learn":[0.4059170464],"iteration":669,"passed_time":143.3098167,"remaining_time":177.5330565,"test":[0.5222055966]}, -{"learn":[0.4058050638],"iteration":670,"passed_time":143.5473802,"remaining_time":177.3484027,"test":[0.5222149749]}, -{"learn":[0.4056596976],"iteration":671,"passed_time":143.7759889,"remaining_time":177.1525577,"test":[0.5221594231]}, -{"learn":[0.4055702172],"iteration":672,"passed_time":143.9796453,"remaining_time":176.9259534,"test":[0.5221519205]}, -{"learn":[0.4054526882],"iteration":673,"passed_time":144.1902858,"remaining_time":176.7079764,"test":[0.5221677639]}, -{"learn":[0.4053703651],"iteration":674,"passed_time":144.3944495,"remaining_time":176.4821049,"test":[0.5221557915]}, -{"learn":[0.4052619744],"iteration":675,"passed_time":144.6467866,"remaining_time":176.315018,"test":[0.5221788583]}, -{"learn":[0.4051706716],"iteration":676,"passed_time":144.8546189,"remaining_time":176.0935766,"test":[0.5222169703]}, -{"learn":[0.4050839379],"iteration":677,"passed_time":145.0660124,"remaining_time":175.876493,"test":[0.522211064]}, -{"learn":[0.4050310102],"iteration":678,"passed_time":145.2776509,"remaining_time":175.6597222,"test":[0.5221965375]}, -{"learn":[0.4049275583],"iteration":679,"passed_time":145.4921491,"remaining_time":175.4464151,"test":[0.5221970164]}, -{"learn":[0.4048025022],"iteration":680,"passed_time":145.7177777,"remaining_time":175.2464904,"test":[0.5221466127]}, -{"learn":[0.4047124407],"iteration":681,"passed_time":145.9219866,"remaining_time":175.0207992,"test":[0.5221535966]}, -{"learn":[0.4045690818],"iteration":682,"passed_time":146.1331153,"remaining_time":174.8034483,"test":[0.5221871192]}, -{"learn":[0.4044315333],"iteration":683,"passed_time":146.3658115,"remaining_time":174.6118453,"test":[0.5221479297]}, -{"learn":[0.4042882801],"iteration":684,"passed_time":146.5858038,"remaining_time":174.4050075,"test":[0.5220778514]}, -{"learn":[0.404154482],"iteration":685,"passed_time":146.860049,"remaining_time":174.2625071,"test":[0.5220881077]}, -{"learn":[0.4039425865],"iteration":686,"passed_time":147.074552,"remaining_time":174.0489239,"test":[0.522129572]}, -{"learn":[0.4038427529],"iteration":687,"passed_time":147.2836436,"remaining_time":173.8289514,"test":[0.522132605]}, -{"learn":[0.4037263068],"iteration":688,"passed_time":147.4908254,"remaining_time":173.6067625,"test":[0.5221517608]}, -{"learn":[0.4036760467],"iteration":689,"passed_time":147.6926819,"remaining_time":173.3783657,"test":[0.5221073034]}, -{"learn":[0.403583212],"iteration":690,"passed_time":147.9012553,"remaining_time":173.1579096,"test":[0.522141704]}, -{"learn":[0.4034757985],"iteration":691,"passed_time":148.1069277,"remaining_time":172.9341005,"test":[0.5221258207]}, -{"learn":[0.4033589826],"iteration":692,"passed_time":148.3422279,"remaining_time":172.7448454,"test":[0.5221208721]}, -{"learn":[0.403220589],"iteration":693,"passed_time":148.600155,"remaining_time":172.5817362,"test":[0.5220776918]}, -{"learn":[0.4030926277],"iteration":694,"passed_time":148.8164159,"remaining_time":172.3700932,"test":[0.5221266987]}, -{"learn":[0.4029039211],"iteration":695,"passed_time":149.0310233,"remaining_time":172.1565269,"test":[0.5221319665]}, -{"learn":[0.4028226545],"iteration":696,"passed_time":149.2492295,"remaining_time":171.9471037,"test":[0.5221535567]}, -{"learn":[0.402755148],"iteration":697,"passed_time":149.4537521,"remaining_time":171.721933,"test":[0.5221580264]}, -{"learn":[0.4025953614],"iteration":698,"passed_time":149.6627615,"remaining_time":171.5019628,"test":[0.5221366756]}, -{"learn":[0.402450814],"iteration":699,"passed_time":149.8704369,"remaining_time":171.2804993,"test":[0.5221112543]}, -{"learn":[0.4023101754],"iteration":700,"passed_time":150.0890848,"remaining_time":171.0715817,"test":[0.5220847555]}, -{"learn":[0.4022780597],"iteration":701,"passed_time":150.2922142,"remaining_time":170.8449956,"test":[0.5220865513]}, -{"learn":[0.402171729],"iteration":702,"passed_time":150.5351074,"remaining_time":170.663557,"test":[0.5220447278]}, -{"learn":[0.4020247253],"iteration":703,"passed_time":150.7484191,"remaining_time":170.4484966,"test":[0.5220336733]}, -{"learn":[0.4019478165],"iteration":704,"passed_time":150.9927499,"remaining_time":170.2684201,"test":[0.5220583364]}, -{"learn":[0.4017364493],"iteration":705,"passed_time":151.212166,"remaining_time":170.0601414,"test":[0.5220547048]}, -{"learn":[0.4016885397],"iteration":706,"passed_time":151.4225735,"remaining_time":169.8417267,"test":[0.5220721845]}, -{"learn":[0.4015977387],"iteration":707,"passed_time":151.6499997,"remaining_time":169.6423725,"test":[0.5220880678]}, -{"learn":[0.401570641],"iteration":708,"passed_time":151.8604156,"remaining_time":169.4239615,"test":[0.5221017163]}, -{"learn":[0.4015012065],"iteration":709,"passed_time":152.0680343,"remaining_time":169.2024607,"test":[0.522074978]}, -{"learn":[0.4013771276],"iteration":710,"passed_time":152.2754725,"remaining_time":168.9807986,"test":[0.5220769734]}, -{"learn":[0.4013263657],"iteration":711,"passed_time":152.4935658,"remaining_time":168.7709689,"test":[0.5220693909]}, -{"learn":[0.4012401074],"iteration":712,"passed_time":152.7012907,"remaining_time":168.5496715,"test":[0.5220553034]}, -{"learn":[0.4011601349],"iteration":713,"passed_time":152.9073379,"remaining_time":168.3265653,"test":[0.5220848752]}, -{"learn":[0.4010814037],"iteration":714,"passed_time":153.1136096,"remaining_time":168.1037532,"test":[0.5222718038]}, -{"learn":[0.400944938],"iteration":715,"passed_time":153.3255004,"remaining_time":167.8871401,"test":[0.5222876073]}, -{"learn":[0.4008109286],"iteration":716,"passed_time":153.5403852,"remaining_time":167.6738098,"test":[0.5222551223]}, -{"learn":[0.4006492141],"iteration":717,"passed_time":153.7554547,"remaining_time":167.4606762,"test":[0.522198932]}, -{"learn":[0.400531421],"iteration":718,"passed_time":153.9723799,"remaining_time":167.2495531,"test":[0.5222487769]}, -{"learn":[0.4003784749],"iteration":719,"passed_time":154.2031363,"remaining_time":167.0533977,"test":[0.5222378821]}, -{"learn":[0.400226374],"iteration":720,"passed_time":154.4232553,"remaining_time":166.8456531,"test":[0.5222573571]}, -{"learn":[0.4001478805],"iteration":721,"passed_time":154.6482571,"remaining_time":166.6431357,"test":[0.5222718836]}, -{"learn":[0.4000406254],"iteration":722,"passed_time":154.8667259,"remaining_time":166.4335353,"test":[0.5222899619]}, -{"learn":[0.3999716927],"iteration":723,"passed_time":155.1077696,"remaining_time":166.2481066,"test":[0.5223036503]}, -{"learn":[0.3998336424],"iteration":724,"passed_time":155.3163561,"remaining_time":166.027829,"test":[0.5223023733]}, -{"learn":[0.3997517683],"iteration":725,"passed_time":155.5307444,"remaining_time":165.8137688,"test":[0.5222796657]}, -{"learn":[0.3997031721],"iteration":726,"passed_time":155.7329776,"remaining_time":165.5867836,"test":[0.5222763533]}, -{"learn":[0.3994946836],"iteration":727,"passed_time":155.9483645,"remaining_time":165.3738151,"test":[0.522278229]}, -{"learn":[0.3994120436],"iteration":728,"passed_time":156.1582118,"remaining_time":165.1549812,"test":[0.5222687309]}, -{"learn":[0.3992452581],"iteration":729,"passed_time":156.3750468,"remaining_time":164.9435425,"test":[0.5222255106]}, -{"learn":[0.3991151311],"iteration":730,"passed_time":156.5936813,"remaining_time":164.7339821,"test":[0.5222417931]}, -{"learn":[0.3989169694],"iteration":731,"passed_time":156.8156456,"remaining_time":164.5278904,"test":[0.5222338514]}, -{"learn":[0.3988042473],"iteration":732,"passed_time":157.0246963,"remaining_time":164.308243,"test":[0.5221923472]}, -{"learn":[0.3987856803],"iteration":733,"passed_time":157.2314275,"remaining_time":164.0862036,"test":[0.5222242336]}, -{"learn":[0.398664269],"iteration":734,"passed_time":157.4558596,"remaining_time":163.8826294,"test":[0.5222261891]}, -{"learn":[0.3985673142],"iteration":735,"passed_time":157.6964397,"remaining_time":163.6957608,"test":[0.5221915889]}, -{"learn":[0.3984359195],"iteration":736,"passed_time":157.9060141,"remaining_time":163.4766468,"test":[0.5222121814]}, -{"learn":[0.3983484991],"iteration":737,"passed_time":158.1206387,"remaining_time":163.2627732,"test":[0.5222113832]}, -{"learn":[0.3983241217],"iteration":738,"passed_time":158.3249152,"remaining_time":163.0382415,"test":[0.5222260294]}, -{"learn":[0.3982504614],"iteration":739,"passed_time":158.5475578,"remaining_time":162.8326269,"test":[0.5222252313]}, -{"learn":[0.39816246],"iteration":740,"passed_time":158.7565127,"remaining_time":162.6129463,"test":[0.5222304592]}, -{"learn":[0.3980996282],"iteration":741,"passed_time":159.0412735,"remaining_time":162.4707349,"test":[0.5222567186]}, -{"learn":[0.3980129473],"iteration":742,"passed_time":159.2630401,"remaining_time":162.2639588,"test":[0.5222550824]}, -{"learn":[0.3979900754],"iteration":743,"passed_time":159.4631843,"remaining_time":162.0351712,"test":[0.522232295]}, -{"learn":[0.3978655475],"iteration":744,"passed_time":159.6717654,"remaining_time":161.8150105,"test":[0.5222211208]}, -{"learn":[0.3977282896],"iteration":745,"passed_time":159.8882712,"remaining_time":161.6028907,"test":[0.5222276258]}, -{"learn":[0.3976407635],"iteration":746,"passed_time":160.0964123,"remaining_time":161.3823272,"test":[0.522253087]}, -{"learn":[0.3974670848],"iteration":747,"passed_time":160.3229802,"remaining_time":161.1803224,"test":[0.5223028522]}, -{"learn":[0.3973722165],"iteration":748,"passed_time":160.5372521,"remaining_time":160.965923,"test":[0.5223244024]}, -{"learn":[0.3972935645],"iteration":749,"passed_time":160.7726673,"remaining_time":160.7726673,"test":[0.5223264776]}, -{"learn":[0.3971188557],"iteration":750,"passed_time":161.0082771,"remaining_time":160.5794934,"test":[0.5223473894]}, -{"learn":[0.3969426415],"iteration":751,"passed_time":161.2238827,"remaining_time":160.3663088,"test":[0.5223455536]}, -{"learn":[0.3967768069],"iteration":752,"passed_time":161.4322937,"remaining_time":160.1459806,"test":[0.522416031]}, -{"learn":[0.396643088],"iteration":753,"passed_time":161.6482718,"remaining_time":159.9331707,"test":[0.5224128782]}, -{"learn":[0.3965085768],"iteration":754,"passed_time":161.8715411,"remaining_time":159.7275472,"test":[0.522426846]}, -{"learn":[0.3964347052],"iteration":755,"passed_time":162.0788692,"remaining_time":159.5061888,"test":[0.5224367831]}, -{"learn":[0.3963336303],"iteration":756,"passed_time":162.2888827,"remaining_time":159.2875031,"test":[0.5224471192]}, -{"learn":[0.3962246849],"iteration":757,"passed_time":162.4995078,"remaining_time":159.069439,"test":[0.5224621246]}, -{"learn":[0.3961006589],"iteration":758,"passed_time":162.7186162,"remaining_time":158.8596767,"test":[0.5224556994]}, -{"learn":[0.3959192945],"iteration":759,"passed_time":162.9493859,"remaining_time":158.6612442,"test":[0.5223899711]}, -{"learn":[0.3958587605],"iteration":760,"passed_time":163.1924839,"remaining_time":158.4746986,"test":[0.5224011852]}, -{"learn":[0.3958067044],"iteration":761,"passed_time":163.42247,"remaining_time":158.2753056,"test":[0.5224501123]}, -{"learn":[0.3956605196],"iteration":762,"passed_time":163.6836505,"remaining_time":158.1059638,"test":[0.5224237332]}, -{"learn":[0.3956133759],"iteration":763,"passed_time":163.9060287,"remaining_time":157.8990015,"test":[0.5224197823]}, -{"learn":[0.3955843238],"iteration":764,"passed_time":164.1046845,"remaining_time":157.6692067,"test":[0.5223828675]}, -{"learn":[0.3954913571],"iteration":765,"passed_time":164.3171052,"remaining_time":157.452683,"test":[0.5224311561]}, -{"learn":[0.3954052044],"iteration":766,"passed_time":164.5317985,"remaining_time":157.238342,"test":[0.5224181062]}, -{"learn":[0.3952515981],"iteration":767,"passed_time":164.7493181,"remaining_time":157.0266938,"test":[0.5223989504]}, -{"learn":[0.3951503911],"iteration":768,"passed_time":164.9537031,"remaining_time":156.8025448,"test":[0.5223784377]}, -{"learn":[0.3950491577],"iteration":769,"passed_time":165.1590578,"remaining_time":156.5793665,"test":[0.5223923656]}, -{"learn":[0.3949205362],"iteration":770,"passed_time":165.3738974,"remaining_time":156.3652026,"test":[0.5223688598]}, -{"learn":[0.3948132283],"iteration":771,"passed_time":165.5875269,"remaining_time":156.1498959,"test":[0.5223993495]}, -{"learn":[0.3946939298],"iteration":772,"passed_time":165.7971437,"remaining_time":155.9308195,"test":[0.5224052558]}, -{"learn":[0.3946074866],"iteration":773,"passed_time":166.0075065,"remaining_time":155.7124673,"test":[0.5224538637]}, -{"learn":[0.3945200398],"iteration":774,"passed_time":166.2217218,"remaining_time":155.4977398,"test":[0.522461526]}, -{"learn":[0.3944517409],"iteration":775,"passed_time":166.4421512,"remaining_time":155.2888112,"test":[0.5224400156]}, -{"learn":[0.394347708],"iteration":776,"passed_time":166.6616416,"remaining_time":155.0789792,"test":[0.52247713]}, -{"learn":[0.39426314],"iteration":777,"passed_time":166.8694324,"remaining_time":154.858265,"test":[0.522503988]}, -{"learn":[0.3941691696],"iteration":778,"passed_time":167.0933716,"remaining_time":154.6525301,"test":[0.5224891423]}, -{"learn":[0.3941472485],"iteration":779,"passed_time":167.3103325,"remaining_time":154.4403069,"test":[0.5224936918]}, -{"learn":[0.3940587452],"iteration":780,"passed_time":167.5542806,"remaining_time":154.2529164,"test":[0.5225751438]}, -{"learn":[0.3939841342],"iteration":781,"passed_time":167.7656995,"remaining_time":154.0355144,"test":[0.5225697562]}, -{"learn":[0.3938982192],"iteration":782,"passed_time":167.970744,"remaining_time":153.8122905,"test":[0.5226010839]}, -{"learn":[0.3937222427],"iteration":783,"passed_time":168.1923053,"remaining_time":153.6041972,"test":[0.5226017624]}, -{"learn":[0.3936012803],"iteration":784,"passed_time":168.417801,"remaining_time":153.3996532,"test":[0.522541701]}, -{"learn":[0.3935053556],"iteration":785,"passed_time":168.62665,"remaining_time":153.179934,"test":[0.5225556288]}, -{"learn":[0.393446327],"iteration":786,"passed_time":168.8363504,"remaining_time":152.9610138,"test":[0.5225448138]}, -{"learn":[0.3932816016],"iteration":787,"passed_time":169.0507547,"remaining_time":152.7463672,"test":[0.522550361]} +{"learn":[0.6887138826],"iteration":0,"passed_time":0.2709717078,"remaining_time":406.1865899,"test":[0.6894516063]}, +{"learn":[0.6843872326],"iteration":1,"passed_time":0.4812678685,"remaining_time":360.4696335,"test":[0.6855813377]}, +{"learn":[0.680262838],"iteration":2,"passed_time":0.6902650401,"remaining_time":344.442255,"test":[0.6821273883]}, +{"learn":[0.6762009847],"iteration":3,"passed_time":0.9019649785,"remaining_time":337.334902,"test":[0.6785411841]}, +{"learn":[0.6723192017],"iteration":4,"passed_time":0.9480610433,"remaining_time":283.4702519,"test":[0.6752506177]}, +{"learn":[0.6682080126],"iteration":5,"passed_time":1.191684141,"remaining_time":296.729351,"test":[0.6727377765]}, +{"learn":[0.6642343721],"iteration":6,"passed_time":1.387762589,"remaining_time":295.9899351,"test":[0.6695440266]}, +{"learn":[0.6604090029],"iteration":7,"passed_time":1.586502997,"remaining_time":295.8828089,"test":[0.6666962384]}, +{"learn":[0.6567664506],"iteration":8,"passed_time":1.873528866,"remaining_time":310.3812821,"test":[0.6635424364]}, +{"learn":[0.6530918882],"iteration":9,"passed_time":2.092919986,"remaining_time":311.8450779,"test":[0.660442949]}, +{"learn":[0.6495994032],"iteration":10,"passed_time":2.31293028,"remaining_time":313.0866533,"test":[0.6580995967]}, +{"learn":[0.645985639],"iteration":11,"passed_time":2.566475852,"remaining_time":318.2430056,"test":[0.6556190011]}, +{"learn":[0.6426158067],"iteration":12,"passed_time":2.796176964,"remaining_time":319.8396265,"test":[0.6533187095]}, +{"learn":[0.6392257964],"iteration":13,"passed_time":3.010518702,"remaining_time":319.5450566,"test":[0.6512931039]}, +{"learn":[0.6358980104],"iteration":14,"passed_time":3.178990817,"remaining_time":314.7200908,"test":[0.6483628261]}, +{"learn":[0.6326337693],"iteration":15,"passed_time":3.389681924,"remaining_time":314.3929985,"test":[0.6462357347]}, +{"learn":[0.6296338495],"iteration":16,"passed_time":3.494064623,"remaining_time":304.8057551,"test":[0.6434692389]}, +{"learn":[0.6265263841],"iteration":17,"passed_time":3.598875359,"remaining_time":296.3074045,"test":[0.6415186602]}, +{"learn":[0.6235141568],"iteration":18,"passed_time":3.815555239,"remaining_time":297.4124899,"test":[0.6387646556]}, +{"learn":[0.6205433947],"iteration":19,"passed_time":4.036901531,"remaining_time":298.7307133,"test":[0.6368736993]}, +{"learn":[0.6175790241],"iteration":20,"passed_time":4.254624922,"remaining_time":299.6471553,"test":[0.63471105]}, +{"learn":[0.6147450182],"iteration":21,"passed_time":4.489168928,"remaining_time":301.5905307,"test":[0.6322352833]}, +{"learn":[0.6119898227],"iteration":22,"passed_time":4.721369101,"remaining_time":303.194007,"test":[0.6300870008]}, +{"learn":[0.6092233761],"iteration":23,"passed_time":4.962412586,"remaining_time":305.188374,"test":[0.6275533277]}, +{"learn":[0.6065009302],"iteration":24,"passed_time":5.195468762,"remaining_time":306.532657,"test":[0.6257390345]}, +{"learn":[0.6037280922],"iteration":25,"passed_time":5.418677564,"remaining_time":307.1973358,"test":[0.6242293983]}, +{"learn":[0.601134849],"iteration":26,"passed_time":5.630134794,"remaining_time":307.1551315,"test":[0.6220503069]}, +{"learn":[0.5985698127],"iteration":27,"passed_time":5.833477005,"remaining_time":306.6742197,"test":[0.6199216991]}, +{"learn":[0.595987979],"iteration":28,"passed_time":6.08015682,"remaining_time":308.4107132,"test":[0.6184262302]}, +{"learn":[0.5935678334],"iteration":29,"passed_time":6.280839172,"remaining_time":307.7611194,"test":[0.6162235133]}, +{"learn":[0.5912109686],"iteration":30,"passed_time":6.493414299,"remaining_time":307.7040518,"test":[0.6141742424]}, +{"learn":[0.588862661],"iteration":31,"passed_time":6.704081361,"remaining_time":307.5497324,"test":[0.6127877619]}, +{"learn":[0.5865620516],"iteration":32,"passed_time":6.921094353,"remaining_time":307.6741035,"test":[0.6108688302]}, +{"learn":[0.5842648229],"iteration":33,"passed_time":7.132885604,"remaining_time":307.553244,"test":[0.6096360351]}, +{"learn":[0.5820779921],"iteration":34,"passed_time":7.346522421,"remaining_time":307.5044385,"test":[0.6077157864]}, +{"learn":[0.5799317814],"iteration":35,"passed_time":7.575778374,"remaining_time":308.0816539,"test":[0.6064106782]}, +{"learn":[0.5777438942],"iteration":36,"passed_time":7.802411309,"remaining_time":308.5115607,"test":[0.6055380919]}, +{"learn":[0.5756289014],"iteration":37,"passed_time":8.031782012,"remaining_time":309.0122448,"test":[0.6043700673]}, +{"learn":[0.5736006422],"iteration":38,"passed_time":8.300426777,"remaining_time":310.9467569,"test":[0.6026869913]}, +{"learn":[0.5715183462],"iteration":39,"passed_time":8.527137671,"remaining_time":311.240525,"test":[0.6015792676]}, +{"learn":[0.5695854836],"iteration":40,"passed_time":8.750204806,"remaining_time":311.3792393,"test":[0.5999814348]}, +{"learn":[0.5676575862],"iteration":41,"passed_time":8.964542157,"remaining_time":311.1976777,"test":[0.5986109974]}, +{"learn":[0.5658429393],"iteration":42,"passed_time":9.178798791,"remaining_time":311.0118567,"test":[0.5973064079]}, +{"learn":[0.5640453538],"iteration":43,"passed_time":9.394826034,"remaining_time":310.8833342,"test":[0.59606619]}, +{"learn":[0.5623314384],"iteration":44,"passed_time":9.494634802,"remaining_time":306.9931919,"test":[0.5948853151]}, +{"learn":[0.5605909535],"iteration":45,"passed_time":9.698282715,"remaining_time":306.5500667,"test":[0.5939087686]}, +{"learn":[0.5587756199],"iteration":46,"passed_time":9.916311482,"remaining_time":306.5617146,"test":[0.5927287318]}, +{"learn":[0.5569407949],"iteration":47,"passed_time":10.1451973,"remaining_time":306.8922183,"test":[0.591221929]}, +{"learn":[0.5552666016],"iteration":48,"passed_time":10.35589735,"remaining_time":306.6613686,"test":[0.5902841331]}, +{"learn":[0.5535532672],"iteration":49,"passed_time":10.58214537,"remaining_time":306.8822157,"test":[0.5892440533]}, +{"learn":[0.5520046847],"iteration":50,"passed_time":10.79009253,"remaining_time":306.5655703,"test":[0.5881410786]}, +{"learn":[0.550447228],"iteration":51,"passed_time":11.02257159,"remaining_time":306.9362244,"test":[0.586657582]}, +{"learn":[0.5489185594],"iteration":52,"passed_time":11.25454778,"remaining_time":307.2703893,"test":[0.5855571215]}, +{"learn":[0.5474294015],"iteration":53,"passed_time":11.48142851,"remaining_time":307.4471412,"test":[0.5845543955]}, +{"learn":[0.5459805469],"iteration":54,"passed_time":11.74401692,"remaining_time":308.5473535,"test":[0.5833092689]}, +{"learn":[0.5445915924],"iteration":55,"passed_time":12.07415412,"remaining_time":311.3406885,"test":[0.5818971675]}, +{"learn":[0.5431495519],"iteration":56,"passed_time":12.33627607,"remaining_time":312.3025678,"test":[0.580668882]}, +{"learn":[0.5417205583],"iteration":57,"passed_time":12.55276516,"remaining_time":312.0877132,"test":[0.5798168083]}, +{"learn":[0.5403852182],"iteration":58,"passed_time":12.77025807,"remaining_time":311.8973199,"test":[0.5785289403]}, +{"learn":[0.5389835864],"iteration":59,"passed_time":12.99168087,"remaining_time":311.8003409,"test":[0.5776724369]}, +{"learn":[0.5377107347],"iteration":60,"passed_time":13.25026497,"remaining_time":312.5759228,"test":[0.5765605627]}, +{"learn":[0.5364798765],"iteration":61,"passed_time":13.46638372,"remaining_time":312.3332224,"test":[0.57530941]}, +{"learn":[0.5352337527],"iteration":62,"passed_time":13.70062368,"remaining_time":312.5047021,"test":[0.5742001298]}, +{"learn":[0.5339944957],"iteration":63,"passed_time":13.9289373,"remaining_time":312.5305306,"test":[0.5732021928]}, +{"learn":[0.5327745717],"iteration":64,"passed_time":14.15199752,"remaining_time":312.4325606,"test":[0.5721310646]}, +{"learn":[0.5315705998],"iteration":65,"passed_time":14.36984238,"remaining_time":312.2174845,"test":[0.5711689249]}, +{"learn":[0.5304691555],"iteration":66,"passed_time":14.57699679,"remaining_time":311.7736777,"test":[0.570040489]}, +{"learn":[0.529349963],"iteration":67,"passed_time":14.77600954,"remaining_time":311.1653774,"test":[0.5690538059]}, +{"learn":[0.528244557],"iteration":68,"passed_time":14.98652533,"remaining_time":310.8075036,"test":[0.5685098612]}, +{"learn":[0.527115698],"iteration":69,"passed_time":15.19194609,"remaining_time":310.3497559,"test":[0.567835218]}, +{"learn":[0.5260443094],"iteration":70,"passed_time":15.39607256,"remaining_time":309.873066,"test":[0.5669510983]}, +{"learn":[0.5249989092],"iteration":71,"passed_time":15.60916788,"remaining_time":309.5818296,"test":[0.5659244275]}, +{"learn":[0.5239809237],"iteration":72,"passed_time":15.81362514,"remaining_time":309.1238777,"test":[0.5654294897]}, +{"learn":[0.5229378476],"iteration":73,"passed_time":16.01629321,"remaining_time":308.6382989,"test":[0.5649020668]}, +{"learn":[0.5219464843],"iteration":74,"passed_time":16.22232411,"remaining_time":308.2241581,"test":[0.564184363]}, +{"learn":[0.5208603056],"iteration":75,"passed_time":16.43201383,"remaining_time":307.8840486,"test":[0.563682561]}, +{"learn":[0.5199107245],"iteration":76,"passed_time":16.6380294,"remaining_time":307.4794264,"test":[0.5628025917]}, +{"learn":[0.5189215269],"iteration":77,"passed_time":16.84611454,"remaining_time":307.1176265,"test":[0.5626036907]}, +{"learn":[0.5179007418],"iteration":78,"passed_time":17.05580016,"remaining_time":306.7885066,"test":[0.5622698611]}, +{"learn":[0.5170257982],"iteration":79,"passed_time":17.26860098,"remaining_time":306.5176674,"test":[0.5612862909]}, +{"learn":[0.5160979797],"iteration":80,"passed_time":17.47352724,"remaining_time":306.1103106,"test":[0.5607771857]}, +{"learn":[0.5152053936],"iteration":81,"passed_time":17.69315491,"remaining_time":305.9621178,"test":[0.5598296523]}, +{"learn":[0.5143571779],"iteration":82,"passed_time":17.89501005,"remaining_time":305.508786,"test":[0.5594330076]}, +{"learn":[0.5135479977],"iteration":83,"passed_time":18.09952268,"remaining_time":305.1062394,"test":[0.5586549232]}, +{"learn":[0.5127352784],"iteration":84,"passed_time":18.30442224,"remaining_time":304.7147938,"test":[0.5578690168]}, +{"learn":[0.5118935599],"iteration":85,"passed_time":18.523207,"remaining_time":304.5559848,"test":[0.5573709662]}, +{"learn":[0.5110973739],"iteration":86,"passed_time":18.73827778,"remaining_time":304.3354771,"test":[0.5569636661]}, +{"learn":[0.5103353109],"iteration":87,"passed_time":18.94039565,"remaining_time":303.9072575,"test":[0.5566546991]}, +{"learn":[0.5094644874],"iteration":88,"passed_time":19.15093143,"remaining_time":303.6175758,"test":[0.5559996507]}, +{"learn":[0.5086953991],"iteration":89,"passed_time":19.36203801,"remaining_time":303.3385956,"test":[0.5555010414]}, +{"learn":[0.5079769143],"iteration":90,"passed_time":19.56663944,"remaining_time":302.9603843,"test":[0.5550565471]}, +{"learn":[0.5072724272],"iteration":91,"passed_time":19.77272244,"remaining_time":302.6086216,"test":[0.5546474911]}, +{"learn":[0.5064938838],"iteration":92,"passed_time":19.98596072,"remaining_time":302.3682444,"test":[0.5539547297]}, +{"learn":[0.505699705],"iteration":93,"passed_time":20.19954694,"remaining_time":302.1336489,"test":[0.5533246238]}, +{"learn":[0.5049644228],"iteration":94,"passed_time":20.40307209,"remaining_time":301.7506978,"test":[0.5528296062]}, +{"learn":[0.5042152484],"iteration":95,"passed_time":20.60621294,"remaining_time":301.3658643,"test":[0.5525477366]}, +{"learn":[0.5035045284],"iteration":96,"passed_time":20.81137859,"remaining_time":301.0140635,"test":[0.5519693514]}, +{"learn":[0.5028365414],"iteration":97,"passed_time":21.01528169,"remaining_time":300.6471931,"test":[0.5516401111]}, +{"learn":[0.5021300471],"iteration":98,"passed_time":21.22875454,"remaining_time":300.4190415,"test":[0.5510557397]}, +{"learn":[0.5014405087],"iteration":99,"passed_time":21.43743303,"remaining_time":300.1240624,"test":[0.5507846453]}, +{"learn":[0.500777487],"iteration":100,"passed_time":21.64187323,"remaining_time":299.7720857,"test":[0.5506433713]}, +{"learn":[0.500143042],"iteration":101,"passed_time":21.84576387,"remaining_time":299.4154696,"test":[0.5502156384]}, +{"learn":[0.4995490058],"iteration":102,"passed_time":22.04958024,"remaining_time":299.0608116,"test":[0.5496516599]}, +{"learn":[0.4989485253],"iteration":103,"passed_time":22.25902587,"remaining_time":298.7846165,"test":[0.5492036537]}, +{"learn":[0.4983499992],"iteration":104,"passed_time":22.46729036,"remaining_time":298.4940005,"test":[0.5488685869]}, +{"learn":[0.4977428103],"iteration":105,"passed_time":22.69578653,"remaining_time":298.4710039,"test":[0.5482614281]}, +{"learn":[0.4971949405],"iteration":106,"passed_time":22.9072734,"remaining_time":298.2227275,"test":[0.5479094403]}, +{"learn":[0.4966445881],"iteration":107,"passed_time":23.1170725,"remaining_time":297.9533789,"test":[0.5476357918]}, +{"learn":[0.4960739521],"iteration":108,"passed_time":23.33062556,"remaining_time":297.7330289,"test":[0.5473086268]}, +{"learn":[0.4954705664],"iteration":109,"passed_time":23.53963563,"remaining_time":297.4553956,"test":[0.5471436875]}, +{"learn":[0.4948636944],"iteration":110,"passed_time":23.74754608,"remaining_time":297.1652388,"test":[0.5466196967]}, +{"learn":[0.4943254382],"iteration":111,"passed_time":23.95409893,"remaining_time":296.859726,"test":[0.5463429353]}, +{"learn":[0.4937686944],"iteration":112,"passed_time":24.16154675,"remaining_time":296.5669499,"test":[0.5462688663]}, +{"learn":[0.4932566907],"iteration":113,"passed_time":24.36451775,"remaining_time":296.2212421,"test":[0.5460389967]}, +{"learn":[0.4927494939],"iteration":114,"passed_time":24.56860941,"remaining_time":295.8915133,"test":[0.5458030612]}, +{"learn":[0.4922379128],"iteration":115,"passed_time":24.76841725,"remaining_time":295.5128403,"test":[0.5455376337]}, +{"learn":[0.4917601378],"iteration":116,"passed_time":24.97640722,"remaining_time":295.2339418,"test":[0.5450453298]}, +{"learn":[0.4912689461],"iteration":117,"passed_time":25.18476599,"remaining_time":294.9605643,"test":[0.5448328601]}, +{"learn":[0.4907662919],"iteration":118,"passed_time":25.3964231,"remaining_time":294.7265571,"test":[0.5444673436]}, +{"learn":[0.4903079554],"iteration":119,"passed_time":25.60763529,"remaining_time":294.4878058,"test":[0.5442106959]}, +{"learn":[0.4896477333],"iteration":120,"passed_time":25.82294566,"remaining_time":294.2962154,"test":[0.5439807864]}, +{"learn":[0.4891970031],"iteration":121,"passed_time":26.03961597,"remaining_time":294.1195967,"test":[0.5434537626]}, +{"learn":[0.4886972014],"iteration":122,"passed_time":26.25479185,"remaining_time":293.9255965,"test":[0.5430699683]}, +{"learn":[0.4882629517],"iteration":123,"passed_time":26.46267936,"remaining_time":293.6503774,"test":[0.5427607618]}, +{"learn":[0.4878229973],"iteration":124,"passed_time":26.67152173,"remaining_time":293.3867391,"test":[0.5426135017]}, +{"learn":[0.4874311637],"iteration":125,"passed_time":26.88789213,"remaining_time":293.2060618,"test":[0.5422171364]}, +{"learn":[0.4870381152],"iteration":126,"passed_time":27.1085779,"remaining_time":293.0714761,"test":[0.5420255385]}, +{"learn":[0.486649715],"iteration":127,"passed_time":27.30937056,"remaining_time":292.7223157,"test":[0.5418352975]}, +{"learn":[0.4862709813],"iteration":128,"passed_time":27.51366021,"remaining_time":292.4126213,"test":[0.5415266498]}, +{"learn":[0.4858681079],"iteration":129,"passed_time":27.73054179,"remaining_time":292.2372481,"test":[0.5410198992]}, +{"learn":[0.4853898575],"iteration":130,"passed_time":27.94094141,"remaining_time":291.9935023,"test":[0.5406753343]}, +{"learn":[0.4849540232],"iteration":131,"passed_time":28.15328548,"remaining_time":291.7704132,"test":[0.5404109445]}, +{"learn":[0.4845096317],"iteration":132,"passed_time":28.36155868,"remaining_time":291.5056445,"test":[0.5400712883]}, +{"learn":[0.4841040116],"iteration":133,"passed_time":28.56621316,"remaining_time":291.2048297,"test":[0.5399532407]}, +{"learn":[0.4836538097],"iteration":134,"passed_time":28.77985618,"remaining_time":290.9963236,"test":[0.5396575231]}, +{"learn":[0.4832326599],"iteration":135,"passed_time":28.98589272,"remaining_time":290.7114535,"test":[0.5393447649]}, +{"learn":[0.4828747908],"iteration":136,"passed_time":29.19996784,"remaining_time":290.5077092,"test":[0.5392009767]}, +{"learn":[0.4824942612],"iteration":137,"passed_time":29.41110554,"remaining_time":290.2748242,"test":[0.539071276]}, +{"learn":[0.4820717908],"iteration":138,"passed_time":29.62343935,"remaining_time":290.0539637,"test":[0.5387476628]}, +{"learn":[0.4817320925],"iteration":139,"passed_time":29.83504515,"remaining_time":289.8261529,"test":[0.5385680373]}, +{"learn":[0.4813567394],"iteration":140,"passed_time":30.0591598,"remaining_time":289.7191359,"test":[0.5384274019]}, +{"learn":[0.4809972329],"iteration":141,"passed_time":30.29225764,"remaining_time":289.6963794,"test":[0.5381121294]}, +{"learn":[0.4806665143],"iteration":142,"passed_time":30.50615559,"remaining_time":289.4884835,"test":[0.5377796167]}, +{"learn":[0.4802733602],"iteration":143,"passed_time":30.71224999,"remaining_time":289.2070207,"test":[0.5376424134]}, +{"learn":[0.4799522024],"iteration":144,"passed_time":30.9186393,"remaining_time":288.9293535,"test":[0.5373759882]}, +{"learn":[0.4796209557],"iteration":145,"passed_time":31.16068889,"remaining_time":288.983375,"test":[0.5372650043]}, +{"learn":[0.4792146489],"iteration":146,"passed_time":31.37104653,"remaining_time":288.7416732,"test":[0.5369387572]}, +{"learn":[0.4789183703],"iteration":147,"passed_time":31.57631731,"remaining_time":288.4539257,"test":[0.536795847]}, +{"learn":[0.478532664],"iteration":148,"passed_time":31.7850941,"remaining_time":288.1990747,"test":[0.5366868585]}, +{"learn":[0.4782636944],"iteration":149,"passed_time":31.99018959,"remaining_time":287.9117063,"test":[0.5365283044]}, +{"learn":[0.4779823644],"iteration":150,"passed_time":32.19737933,"remaining_time":287.6441372,"test":[0.5363033834]}, +{"learn":[0.4776756799],"iteration":151,"passed_time":32.40438965,"remaining_time":287.3757713,"test":[0.5361123842]}, +{"learn":[0.4773818311],"iteration":152,"passed_time":32.61279342,"remaining_time":287.1204754,"test":[0.5359007526]}, +{"learn":[0.4770647406],"iteration":153,"passed_time":32.82011514,"remaining_time":286.8563311,"test":[0.5357652652]}, +{"learn":[0.4767061849],"iteration":154,"passed_time":33.03409798,"remaining_time":286.6507212,"test":[0.5357864164]}, +{"learn":[0.4764295561],"iteration":155,"passed_time":33.24423208,"remaining_time":286.4118456,"test":[0.535719012]}, +{"learn":[0.4760174389],"iteration":156,"passed_time":33.45835241,"remaining_time":286.2074349,"test":[0.5356127372]}, +{"learn":[0.4756622109],"iteration":157,"passed_time":33.66739053,"remaining_time":285.9597347,"test":[0.5353806727]}, +{"learn":[0.4753690488],"iteration":158,"passed_time":33.87666268,"remaining_time":285.7144947,"test":[0.535390969]}, +{"learn":[0.4751120698],"iteration":159,"passed_time":34.08101802,"remaining_time":285.4285259,"test":[0.5351500849]}, +{"learn":[0.4748108259],"iteration":160,"passed_time":34.29088842,"remaining_time":285.1894385,"test":[0.5351811332]}, +{"learn":[0.4745127514],"iteration":161,"passed_time":34.49662137,"remaining_time":284.9165394,"test":[0.5351523197]}, +{"learn":[0.4742213852],"iteration":162,"passed_time":34.70675138,"remaining_time":284.6805313,"test":[0.5348433527]}, +{"learn":[0.4739676284],"iteration":163,"passed_time":34.92025287,"remaining_time":284.4723038,"test":[0.5345701432]}, +{"learn":[0.4736240212],"iteration":164,"passed_time":35.1389719,"remaining_time":284.3062272,"test":[0.5343011639]}, +{"learn":[0.4733876956],"iteration":165,"passed_time":35.35597043,"remaining_time":284.1256901,"test":[0.5340034509]}, +{"learn":[0.473145718],"iteration":166,"passed_time":35.56449822,"remaining_time":283.8771026,"test":[0.5338239053]}, +{"learn":[0.4729062759],"iteration":167,"passed_time":35.76784745,"remaining_time":283.5879334,"test":[0.533652421]}, +{"learn":[0.4726251572],"iteration":168,"passed_time":35.97647394,"remaining_time":283.3413421,"test":[0.5334977778]}, +{"learn":[0.4723726153],"iteration":169,"passed_time":36.1880246,"remaining_time":283.1180748,"test":[0.5333416182]}, +{"learn":[0.4720965675],"iteration":170,"passed_time":36.39213584,"remaining_time":282.8371259,"test":[0.5331617134]}, +{"learn":[0.4718569669],"iteration":171,"passed_time":36.59975089,"remaining_time":282.5841231,"test":[0.5330645775]}, +{"learn":[0.4715016333],"iteration":172,"passed_time":36.80735544,"remaining_time":282.3315645,"test":[0.5329321631]}, +{"learn":[0.4712788301],"iteration":173,"passed_time":37.00966875,"remaining_time":282.0391998,"test":[0.5327248415]}, +{"learn":[0.4709921123],"iteration":174,"passed_time":37.22648497,"remaining_time":281.8576719,"test":[0.532655721]}, +{"learn":[0.4707643438],"iteration":175,"passed_time":37.44245465,"remaining_time":281.6693747,"test":[0.5324228584]}, +{"learn":[0.4704056825],"iteration":176,"passed_time":37.65771811,"remaining_time":281.4754862,"test":[0.5324790887]}, +{"learn":[0.4701234017],"iteration":177,"passed_time":37.86879276,"remaining_time":281.2502473,"test":[0.5322301431]}, +{"learn":[0.4698584994],"iteration":178,"passed_time":38.0817587,"remaining_time":281.0391242,"test":[0.5321801385]}, +{"learn":[0.469572838],"iteration":179,"passed_time":38.29425151,"remaining_time":280.8245111,"test":[0.5319024992]}, +{"learn":[0.4693260536],"iteration":180,"passed_time":38.50509667,"remaining_time":280.5979144,"test":[0.531768249]}, +{"learn":[0.4690975985],"iteration":181,"passed_time":38.71351251,"remaining_time":280.3538983,"test":[0.5317026005]}, +{"learn":[0.4688616426],"iteration":182,"passed_time":38.92411057,"remaining_time":280.1259761,"test":[0.5314705361]}, +{"learn":[0.4685304487],"iteration":183,"passed_time":39.13897961,"remaining_time":279.928789,"test":[0.5313124609]}, +{"learn":[0.4682865695],"iteration":184,"passed_time":39.34862358,"remaining_time":279.6942703,"test":[0.5312031132]}, +{"learn":[0.4680025456],"iteration":185,"passed_time":39.56794085,"remaining_time":279.5283563,"test":[0.5313068738]}, +{"learn":[0.4677844436],"iteration":186,"passed_time":39.78126116,"remaining_time":279.3197642,"test":[0.5310851454]}, +{"learn":[0.4674938169],"iteration":187,"passed_time":39.98899122,"remaining_time":279.072109,"test":[0.5309420357]}, +{"learn":[0.4672483003],"iteration":188,"passed_time":40.19552129,"remaining_time":278.8165524,"test":[0.5308134125]}, +{"learn":[0.4669966034],"iteration":189,"passed_time":40.40971102,"remaining_time":278.6143234,"test":[0.5305983089]}, +{"learn":[0.466680411],"iteration":190,"passed_time":40.6337847,"remaining_time":278.4797077,"test":[0.5304503703]}, +{"learn":[0.4664587698],"iteration":191,"passed_time":40.84338149,"remaining_time":278.2455364,"test":[0.5304046358]}, +{"learn":[0.4662118798],"iteration":192,"passed_time":41.05408203,"remaining_time":278.0190944,"test":[0.5304474969]}, +{"learn":[0.4660336056],"iteration":193,"passed_time":41.26704205,"remaining_time":277.8080254,"test":[0.5302670334]}, +{"learn":[0.465759565],"iteration":194,"passed_time":41.47368966,"remaining_time":277.5546923,"test":[0.530215632]}, +{"learn":[0.4655335925],"iteration":195,"passed_time":41.6835718,"remaining_time":277.3233553,"test":[0.5302616458]}, +{"learn":[0.4653786656],"iteration":196,"passed_time":41.8900307,"remaining_time":277.0695939,"test":[0.5301575261]}, +{"learn":[0.4651596656],"iteration":197,"passed_time":42.09928896,"remaining_time":276.8347183,"test":[0.5299640126]}, +{"learn":[0.4649593117],"iteration":198,"passed_time":42.30810561,"remaining_time":276.597213,"test":[0.5298924179]}, +{"learn":[0.464720345],"iteration":199,"passed_time":42.52222358,"remaining_time":276.3944533,"test":[0.529908381]}, +{"learn":[0.4645237944],"iteration":200,"passed_time":42.73362798,"remaining_time":276.1740435,"test":[0.5297052099]}, +{"learn":[0.4642531344],"iteration":201,"passed_time":42.94294582,"remaining_time":275.9403152,"test":[0.5296109075]}, +{"learn":[0.4640300671],"iteration":202,"passed_time":43.14933337,"remaining_time":275.6881053,"test":[0.5295602644]}, +{"learn":[0.4638038305],"iteration":203,"passed_time":43.35104268,"remaining_time":275.4066241,"test":[0.5294535905]}, +{"learn":[0.4636120866],"iteration":204,"passed_time":43.56372798,"remaining_time":275.1952572,"test":[0.5293538607]}, +{"learn":[0.4633936149],"iteration":205,"passed_time":43.81028628,"remaining_time":275.1966526,"test":[0.5292528538]}, +{"learn":[0.4631845982],"iteration":206,"passed_time":44.02285636,"remaining_time":274.9833491,"test":[0.5291430272]}, +{"learn":[0.4630174694],"iteration":207,"passed_time":44.23699264,"remaining_time":274.7797812,"test":[0.5291064317]}, +{"learn":[0.462711947],"iteration":208,"passed_time":44.45141119,"remaining_time":274.5778557,"test":[0.5291368016]}, +{"learn":[0.4625003948],"iteration":209,"passed_time":44.66107522,"remaining_time":274.3466049,"test":[0.5290518377]}, +{"learn":[0.4623027349],"iteration":210,"passed_time":44.87013903,"remaining_time":274.111892,"test":[0.5290010349]}, +{"learn":[0.4620509853],"iteration":211,"passed_time":45.07393085,"remaining_time":273.8453912,"test":[0.5288788369]}, +{"learn":[0.4618490996],"iteration":212,"passed_time":45.28329339,"remaining_time":273.6131389,"test":[0.5288099957]}, +{"learn":[0.4616663355],"iteration":213,"passed_time":45.48908799,"remaining_time":273.3596596,"test":[0.5287029627]}, +{"learn":[0.4614388839],"iteration":214,"passed_time":45.69611129,"remaining_time":273.1139675,"test":[0.5286111347]}, +{"learn":[0.4611913601],"iteration":215,"passed_time":45.90556224,"remaining_time":272.8830644,"test":[0.5286519205]}, +{"learn":[0.4609538195],"iteration":216,"passed_time":46.11971305,"remaining_time":272.6801468,"test":[0.5285690318]}, +{"learn":[0.4608259375],"iteration":217,"passed_time":46.32224231,"remaining_time":272.4087827,"test":[0.5284839083]}, +{"learn":[0.4606163398],"iteration":218,"passed_time":46.53182329,"remaining_time":272.1792951,"test":[0.5283307018]}, +{"learn":[0.4604787385],"iteration":219,"passed_time":46.7388632,"remaining_time":271.9352041,"test":[0.5283035245]}, +{"learn":[0.4602659714],"iteration":220,"passed_time":46.95016202,"remaining_time":271.716096,"test":[0.5281441723]}, +{"learn":[0.4600530988],"iteration":221,"passed_time":47.17246173,"remaining_time":271.5603878,"test":[0.5279369704]}, +{"learn":[0.4598740322],"iteration":222,"passed_time":47.3801675,"remaining_time":271.3205108,"test":[0.5279410809]}, +{"learn":[0.4596930112],"iteration":223,"passed_time":47.59418012,"remaining_time":271.1168475,"test":[0.5277739466]}, +{"learn":[0.4594828325],"iteration":224,"passed_time":47.80490846,"remaining_time":270.8944813,"test":[0.5278144531]}, +{"learn":[0.4593606024],"iteration":225,"passed_time":48.06031648,"remaining_time":270.9240849,"test":[0.5277773388]}, +{"learn":[0.4590814381],"iteration":226,"passed_time":48.26964308,"remaining_time":270.6927561,"test":[0.5276908185]}, +{"learn":[0.4589115097],"iteration":227,"passed_time":48.47933142,"remaining_time":270.4636384,"test":[0.5275789167]}, +{"learn":[0.4587867442],"iteration":228,"passed_time":48.68742195,"remaining_time":270.2258223,"test":[0.5274718039]}, +{"learn":[0.4586156537],"iteration":229,"passed_time":48.89227759,"remaining_time":269.9704023,"test":[0.5273879175]}, +{"learn":[0.4584218498],"iteration":230,"passed_time":49.10606836,"remaining_time":269.7645054,"test":[0.5272032237]}, +{"learn":[0.4582468505],"iteration":231,"passed_time":49.31225612,"remaining_time":269.5169861,"test":[0.5271434416]}, +{"learn":[0.4580919236],"iteration":232,"passed_time":49.51747662,"remaining_time":269.2645617,"test":[0.5271729336]}, +{"learn":[0.457960344],"iteration":233,"passed_time":49.72706406,"remaining_time":269.0361671,"test":[0.5271932068]}, +{"learn":[0.4577696566],"iteration":234,"passed_time":49.94493196,"remaining_time":268.8525061,"test":[0.5271494278]}, +{"learn":[0.4575786522],"iteration":235,"passed_time":50.17686166,"remaining_time":268.7438692,"test":[0.5270163749]}, +{"learn":[0.4574154851],"iteration":236,"passed_time":50.39240161,"remaining_time":268.5468491,"test":[0.5268986864]}, +{"learn":[0.4572041442],"iteration":237,"passed_time":50.60874223,"remaining_time":268.3539189,"test":[0.5268795306]}, +{"learn":[0.4570666486],"iteration":238,"passed_time":50.82210373,"remaining_time":268.1450745,"test":[0.5269485713]}, +{"learn":[0.4568047043],"iteration":239,"passed_time":51.04091064,"remaining_time":267.9647809,"test":[0.5269823733]}, +{"learn":[0.4566409033],"iteration":240,"passed_time":51.27791298,"remaining_time":267.8792218,"test":[0.5268643656]}, +{"learn":[0.456427608],"iteration":241,"passed_time":51.49337928,"remaining_time":267.6804592,"test":[0.5268694739]}, +{"learn":[0.4562375545],"iteration":242,"passed_time":51.70209357,"remaining_time":267.4466322,"test":[0.5267796412]}, +{"learn":[0.4559405892],"iteration":243,"passed_time":51.92048349,"remaining_time":267.2628167,"test":[0.5267057317]}, +{"learn":[0.4558015089],"iteration":244,"passed_time":52.17318236,"remaining_time":267.2544647,"test":[0.5267355429]}, +{"learn":[0.4556457896],"iteration":245,"passed_time":52.38097587,"remaining_time":267.0152185,"test":[0.5266087555]}, +{"learn":[0.4554681492],"iteration":246,"passed_time":52.58694196,"remaining_time":266.7669566,"test":[0.5265850901]}, +{"learn":[0.4552844343],"iteration":247,"passed_time":52.796665,"remaining_time":266.5380024,"test":[0.5266133848]}, +{"learn":[0.4549960262],"iteration":248,"passed_time":53.0126355,"remaining_time":266.3405904,"test":[0.5264907478]}, +{"learn":[0.4547092027],"iteration":249,"passed_time":53.22727449,"remaining_time":266.1363725,"test":[0.5264365928]}, +{"learn":[0.4544524878],"iteration":250,"passed_time":53.43951392,"remaining_time":265.920131,"test":[0.526419153]}, +{"learn":[0.4542734212],"iteration":251,"passed_time":53.6497681,"remaining_time":265.6940896,"test":[0.5263015844]}, +{"learn":[0.4541751195],"iteration":252,"passed_time":53.85411327,"remaining_time":265.4390484,"test":[0.5262358959]}, +{"learn":[0.4540318662],"iteration":253,"passed_time":54.05961133,"remaining_time":265.1900619,"test":[0.526113259]}, +{"learn":[0.453824223],"iteration":254,"passed_time":54.26768519,"remaining_time":264.9539924,"test":[0.5261479788]}, +{"learn":[0.4536671832],"iteration":255,"passed_time":54.47653799,"remaining_time":264.7219268,"test":[0.5260264992]}, +{"learn":[0.4534658785],"iteration":256,"passed_time":54.6861137,"remaining_time":264.4935383,"test":[0.525990622]}, +{"learn":[0.4533309183],"iteration":257,"passed_time":54.89690643,"remaining_time":264.2711542,"test":[0.5258837885]}, +{"learn":[0.4531105449],"iteration":258,"passed_time":55.12474139,"remaining_time":264.1305176,"test":[0.5257997824]}, +{"learn":[0.4528481781],"iteration":259,"passed_time":55.33791801,"remaining_time":263.9193013,"test":[0.525747064]}, +{"learn":[0.4526748163],"iteration":260,"passed_time":55.5506005,"remaining_time":263.7057242,"test":[0.5256948246]}, +{"learn":[0.4525394863],"iteration":261,"passed_time":55.75656623,"remaining_time":263.460416,"test":[0.5257121047]}, +{"learn":[0.4523340087],"iteration":262,"passed_time":55.97008438,"remaining_time":263.2509292,"test":[0.5255855966]}, +{"learn":[0.4521074552],"iteration":263,"passed_time":56.18881432,"remaining_time":263.0658125,"test":[0.5255532313]}, +{"learn":[0.4519561201],"iteration":264,"passed_time":56.42632833,"remaining_time":262.967983,"test":[0.5254442029]}, +{"learn":[0.4517016766],"iteration":265,"passed_time":56.63692958,"remaining_time":262.7442523,"test":[0.5253342167]}, +{"learn":[0.451566241],"iteration":266,"passed_time":56.84455905,"remaining_time":262.5068963,"test":[0.5252585913]}, +{"learn":[0.4513972634],"iteration":267,"passed_time":57.05662631,"remaining_time":262.2901628,"test":[0.5251232236]}, +{"learn":[0.4511307236],"iteration":268,"passed_time":57.27485776,"remaining_time":262.1016725,"test":[0.5250108829]}, +{"learn":[0.4510253437],"iteration":269,"passed_time":57.48141717,"remaining_time":261.8597893,"test":[0.5251278929]}, +{"learn":[0.4507912894],"iteration":270,"passed_time":57.68800982,"remaining_time":261.6183176,"test":[0.5251599788]}, +{"learn":[0.4506708024],"iteration":271,"passed_time":57.89655802,"remaining_time":261.3859311,"test":[0.5251249397]}, +{"learn":[0.4504390195],"iteration":272,"passed_time":58.10662889,"remaining_time":261.1605628,"test":[0.5250322735]}, +{"learn":[0.4502752713],"iteration":273,"passed_time":58.31281768,"remaining_time":260.917936,"test":[0.5249261185]}, +{"learn":[0.4500877532],"iteration":274,"passed_time":58.52728024,"remaining_time":260.7124301,"test":[0.524854404]}, +{"learn":[0.4498927872],"iteration":275,"passed_time":58.73892536,"remaining_time":260.4943646,"test":[0.5247326849]}, +{"learn":[0.4497752582],"iteration":276,"passed_time":58.94859017,"remaining_time":260.2676021,"test":[0.5246913403]}, +{"learn":[0.4497071179],"iteration":277,"passed_time":59.15586067,"remaining_time":260.0304379,"test":[0.5246368261]}, +{"learn":[0.4496092387],"iteration":278,"passed_time":59.36820244,"remaining_time":259.8156817,"test":[0.52460047]}, +{"learn":[0.4494378841],"iteration":279,"passed_time":59.57879042,"remaining_time":259.5933011,"test":[0.5244897655]}, +{"learn":[0.4490960201],"iteration":280,"passed_time":59.80479712,"remaining_time":259.4378921,"test":[0.5244507355]}, +{"learn":[0.4489082908],"iteration":281,"passed_time":60.01529677,"remaining_time":259.2150052,"test":[0.5244589166]}, +{"learn":[0.4487697387],"iteration":282,"passed_time":60.22615668,"remaining_time":258.993755,"test":[0.5244354508]}, +{"learn":[0.448658654],"iteration":283,"passed_time":60.45333298,"remaining_time":258.8424398,"test":[0.5244328568]}, +{"learn":[0.4484694984],"iteration":284,"passed_time":60.69825084,"remaining_time":258.7662272,"test":[0.5243957424]}, +{"learn":[0.4483209629],"iteration":285,"passed_time":60.90661667,"remaining_time":258.5336806,"test":[0.5243399113]}, +{"learn":[0.4481748045],"iteration":286,"passed_time":61.11545778,"remaining_time":258.3033111,"test":[0.5242596964]}, +{"learn":[0.4479443949],"iteration":287,"passed_time":61.32241301,"remaining_time":258.0651548,"test":[0.5241995951]}, +{"learn":[0.4477727762],"iteration":288,"passed_time":61.53482438,"remaining_time":257.8500773,"test":[0.5241163073]}, +{"learn":[0.4476346995],"iteration":289,"passed_time":61.74599449,"remaining_time":257.6298391,"test":[0.5240826649]}, +{"learn":[0.4475678797],"iteration":290,"passed_time":61.95371881,"remaining_time":257.3953472,"test":[0.5240546097]}, +{"learn":[0.4474216156],"iteration":291,"passed_time":62.17758695,"remaining_time":257.2278255,"test":[0.5239532436]}, +{"learn":[0.4472752986],"iteration":292,"passed_time":62.39186189,"remaining_time":257.0204004,"test":[0.5238582229]}, +{"learn":[0.4470165765],"iteration":293,"passed_time":62.6098058,"remaining_time":256.8279789,"test":[0.5238159604]}, +{"learn":[0.4468633927],"iteration":294,"passed_time":62.81663876,"remaining_time":256.589999,"test":[0.5237865882]}, +{"learn":[0.4466839564],"iteration":295,"passed_time":63.02818689,"remaining_time":256.3714089,"test":[0.5237659159]}, +{"learn":[0.4465561271],"iteration":296,"passed_time":63.23292248,"remaining_time":256.1252719,"test":[0.5237495137]}, +{"learn":[0.4463075468],"iteration":297,"passed_time":63.44661637,"remaining_time":255.9155466,"test":[0.5237665943]}, +{"learn":[0.4461113659],"iteration":298,"passed_time":63.65738295,"remaining_time":255.6940365,"test":[0.5237160709]}, +{"learn":[0.4460250547],"iteration":299,"passed_time":63.85925155,"remaining_time":255.4370062,"test":[0.5237108829]}, +{"learn":[0.4458235916],"iteration":300,"passed_time":64.07181198,"remaining_time":255.2229321,"test":[0.5236443165]}, +{"learn":[0.4456807609],"iteration":301,"passed_time":64.28084801,"remaining_time":254.9948871,"test":[0.5236073219]}, +{"learn":[0.4454037096],"iteration":302,"passed_time":64.48347145,"remaining_time":254.7416348,"test":[0.5235901215]}, +{"learn":[0.4453104787],"iteration":303,"passed_time":64.71915796,"remaining_time":254.6187925,"test":[0.5235915183]}, +{"learn":[0.4451840757],"iteration":304,"passed_time":64.94718634,"remaining_time":254.4652055,"test":[0.5234947816]}, +{"learn":[0.4450407168],"iteration":305,"passed_time":65.16058531,"remaining_time":254.2540486,"test":[0.5234582658]}, +{"learn":[0.4449429433],"iteration":306,"passed_time":65.37316877,"remaining_time":254.039708,"test":[0.5234427417]}, +{"learn":[0.4447876466],"iteration":307,"passed_time":65.58203646,"remaining_time":253.8109982,"test":[0.5233568997]}, +{"learn":[0.4447108434],"iteration":308,"passed_time":65.78172576,"remaining_time":253.5470401,"test":[0.5233460448]}, +{"learn":[0.4446476155],"iteration":309,"passed_time":65.98720391,"remaining_time":253.3057182,"test":[0.5232778023]}, +{"learn":[0.4445678542],"iteration":310,"passed_time":66.20127353,"remaining_time":253.0974734,"test":[0.5232615997]}, +{"learn":[0.4443813398],"iteration":311,"passed_time":66.41032892,"remaining_time":252.8700986,"test":[0.5232866219]}, +{"learn":[0.4442457457],"iteration":312,"passed_time":66.63421275,"remaining_time":252.6990752,"test":[0.5232186588]}, +{"learn":[0.4441295373],"iteration":313,"passed_time":66.83849616,"remaining_time":252.4536829,"test":[0.523270978]}, +{"learn":[0.4439701205],"iteration":314,"passed_time":67.05142249,"remaining_time":252.2410656,"test":[0.52324412]}, +{"learn":[0.4436869418],"iteration":315,"passed_time":67.25814812,"remaining_time":252.0052132,"test":[0.5232147079]}, +{"learn":[0.4436074446],"iteration":316,"passed_time":67.46084867,"remaining_time":251.7545236,"test":[0.523215945]}, +{"learn":[0.4433314497],"iteration":317,"passed_time":67.6775057,"remaining_time":251.5560118,"test":[0.5232299128]}, +{"learn":[0.4431709765],"iteration":318,"passed_time":67.90023717,"remaining_time":251.3798749,"test":[0.5231812252]}, +{"learn":[0.4430649099],"iteration":319,"passed_time":68.1074843,"remaining_time":251.1463484,"test":[0.5232248046]}, +{"learn":[0.4429950792],"iteration":320,"passed_time":68.33004492,"remaining_time":250.9692304,"test":[0.5231860141]}, +{"learn":[0.4428991809],"iteration":321,"passed_time":68.52940843,"remaining_time":250.7069662,"test":[0.5232448384]}, +{"learn":[0.4428329157],"iteration":322,"passed_time":68.73113042,"remaining_time":250.4536858,"test":[0.5232055291]}, +{"learn":[0.4427574859],"iteration":323,"passed_time":68.94711508,"remaining_time":250.2524918,"test":[0.5232274784]}, +{"learn":[0.4426940995],"iteration":324,"passed_time":69.1489613,"remaining_time":250.0000909,"test":[0.5231533694]}, +{"learn":[0.4426259855],"iteration":325,"passed_time":69.36701291,"remaining_time":249.8063594,"test":[0.5231688138]}, +{"learn":[0.4425582941],"iteration":326,"passed_time":69.58231729,"remaining_time":249.6026244,"test":[0.5231483011]}, +{"learn":[0.4424613393],"iteration":327,"passed_time":69.78801368,"remaining_time":249.3644879,"test":[0.5231146189]}, +{"learn":[0.4422837782],"iteration":328,"passed_time":70.0095777,"remaining_time":249.1830258,"test":[0.5230951039]}, +{"learn":[0.442216166],"iteration":329,"passed_time":70.21473076,"remaining_time":248.9431363,"test":[0.5230877608]}, +{"learn":[0.4421165965],"iteration":330,"passed_time":70.44316742,"remaining_time":248.7856879,"test":[0.5230616611]}, +{"learn":[0.4419779124],"iteration":331,"passed_time":70.65211461,"remaining_time":248.5592466,"test":[0.5230415475]}, +{"learn":[0.4418364022],"iteration":332,"passed_time":70.85926807,"remaining_time":248.3266241,"test":[0.5229808076]}, +{"learn":[0.4417645907],"iteration":333,"passed_time":71.08665167,"remaining_time":248.164778,"test":[0.5229518744]}, +{"learn":[0.441594768],"iteration":334,"passed_time":71.3228679,"remaining_time":248.033257,"test":[0.5229005529]}, +{"learn":[0.4413897658],"iteration":335,"passed_time":71.56728723,"remaining_time":247.9295308,"test":[0.5227853786]}, +{"learn":[0.4412564167],"iteration":336,"passed_time":71.77254583,"remaining_time":247.6898243,"test":[0.522779153]}, +{"learn":[0.4411897553],"iteration":337,"passed_time":71.98223131,"remaining_time":247.4655408,"test":[0.5227785145]}, +{"learn":[0.4409984605],"iteration":338,"passed_time":72.19016694,"remaining_time":247.2353505,"test":[0.5227113495]}, +{"learn":[0.4408125798],"iteration":339,"passed_time":72.39699813,"remaining_time":247.001523,"test":[0.5227014523]}, +{"learn":[0.4407580147],"iteration":340,"passed_time":72.59802833,"remaining_time":246.7481373,"test":[0.5226840126]}, +{"learn":[0.4406348074],"iteration":341,"passed_time":72.80476244,"remaining_time":246.5143711,"test":[0.5226773479]}, +{"learn":[0.4405851283],"iteration":342,"passed_time":73.02477827,"remaining_time":246.325564,"test":[0.5226805406]}, +{"learn":[0.4404759716],"iteration":343,"passed_time":73.25328962,"remaining_time":246.1651244,"test":[0.5226501307]}, +{"learn":[0.440350044],"iteration":344,"passed_time":73.46869986,"remaining_time":245.96043,"test":[0.5226338483]}, +{"learn":[0.4402949506],"iteration":345,"passed_time":73.67710839,"remaining_time":245.732321,"test":[0.5226316534]}, +{"learn":[0.4401106811],"iteration":346,"passed_time":73.89275665,"remaining_time":245.5283816,"test":[0.5226565559]}, +{"learn":[0.4397903685],"iteration":347,"passed_time":74.13123264,"remaining_time":245.3999425,"test":[0.5226600678]}, +{"learn":[0.4397325812],"iteration":348,"passed_time":74.34158508,"remaining_time":245.1781216,"test":[0.5226367217]}, +{"learn":[0.4396825324],"iteration":349,"passed_time":74.54455888,"remaining_time":244.932122,"test":[0.5225947386]}, +{"learn":[0.4396240584],"iteration":350,"passed_time":74.74888776,"remaining_time":244.6908035,"test":[0.5226150517]}, +{"learn":[0.4394675469],"iteration":351,"passed_time":74.95711145,"remaining_time":244.4623976,"test":[0.5225921446]}, +{"learn":[0.4394047943],"iteration":352,"passed_time":75.16113202,"remaining_time":244.2204488,"test":[0.5225689182]}, +{"learn":[0.4393420154],"iteration":353,"passed_time":75.4005504,"remaining_time":244.0933072,"test":[0.5225403441]}, +{"learn":[0.4392638916],"iteration":354,"passed_time":75.61271149,"remaining_time":243.8776187,"test":[0.5225377102]}, +{"learn":[0.4391883298],"iteration":355,"passed_time":75.81826244,"remaining_time":243.6407085,"test":[0.5225459711]}, +{"learn":[0.4391398656],"iteration":356,"passed_time":76.04673148,"remaining_time":243.4773504,"test":[0.5225433372]}, +{"learn":[0.4389915942],"iteration":357,"passed_time":76.25845343,"remaining_time":243.2602062,"test":[0.5225567462]}, +{"learn":[0.4387960208],"iteration":358,"passed_time":76.46404869,"remaining_time":243.0236199,"test":[0.5225509596]}, +{"learn":[0.4385573446],"iteration":359,"passed_time":76.67777651,"remaining_time":242.812959,"test":[0.5225457317]}, +{"learn":[0.4385087747],"iteration":360,"passed_time":76.88942245,"remaining_time":242.5957124,"test":[0.5225362336]}, +{"learn":[0.4383729694],"iteration":361,"passed_time":77.10134137,"remaining_time":242.3793549,"test":[0.5225390271]}, +{"learn":[0.4383069155],"iteration":362,"passed_time":77.40086392,"remaining_time":242.4374167,"test":[0.5225502413]}, +{"learn":[0.4381989737],"iteration":363,"passed_time":77.61697212,"remaining_time":242.2331877,"test":[0.5225557086]}, +{"learn":[0.4380124328],"iteration":364,"passed_time":77.82561627,"remaining_time":242.0056835,"test":[0.5225313648]}, +{"learn":[0.4378701304],"iteration":365,"passed_time":78.04539614,"remaining_time":241.8127848,"test":[0.5224742965]}, +{"learn":[0.4378172555],"iteration":366,"passed_time":78.25076257,"remaining_time":241.5752425,"test":[0.5224713034]}, +{"learn":[0.4377449158],"iteration":367,"passed_time":78.46863212,"remaining_time":241.3763358,"test":[0.5224824377]}, +{"learn":[0.4374820207],"iteration":368,"passed_time":78.68226434,"remaining_time":241.1643387,"test":[0.52239045]}, +{"learn":[0.437295876],"iteration":369,"passed_time":78.89594754,"remaining_time":240.9524884,"test":[0.5223451146]}, +{"learn":[0.4372092215],"iteration":370,"passed_time":79.1028692,"remaining_time":240.7200521,"test":[0.5222921169]}, +{"learn":[0.4370238163],"iteration":371,"passed_time":79.3161498,"remaining_time":240.5070349,"test":[0.5222611883]}, +{"learn":[0.436981902],"iteration":372,"passed_time":79.51570857,"remaining_time":240.2525565,"test":[0.5222393587]}, +{"learn":[0.4368016734],"iteration":373,"passed_time":79.72802001,"remaining_time":240.0367661,"test":[0.52223014]}, +{"learn":[0.4366709653],"iteration":374,"passed_time":79.93688112,"remaining_time":239.8106433,"test":[0.5222536058]}, +{"learn":[0.4365566057],"iteration":375,"passed_time":80.18360679,"remaining_time":239.6978033,"test":[0.5223180172]}, +{"learn":[0.4363945214],"iteration":376,"passed_time":80.40662126,"remaining_time":239.5136225,"test":[0.522292955]}, +{"learn":[0.4363409599],"iteration":377,"passed_time":80.60807718,"remaining_time":239.265245,"test":[0.5222633433]}, +{"learn":[0.4362627305],"iteration":378,"passed_time":80.81221507,"remaining_time":239.0250477,"test":[0.5222620663]}, +{"learn":[0.4361927678],"iteration":379,"passed_time":81.02525644,"remaining_time":238.8112821,"test":[0.5223006971]}, +{"learn":[0.4361403419],"iteration":380,"passed_time":81.22817781,"remaining_time":238.5677978,"test":[0.5223303487]}, +{"learn":[0.4359903539],"iteration":381,"passed_time":81.4731724,"remaining_time":238.4476616,"test":[0.5222801845]}, +{"learn":[0.4357839519],"iteration":382,"passed_time":81.68409286,"remaining_time":238.227498,"test":[0.5223078007]}, +{"learn":[0.435723814],"iteration":383,"passed_time":81.88412755,"remaining_time":237.9757457,"test":[0.5223389688]}, +{"learn":[0.4356793379],"iteration":384,"passed_time":82.08456266,"remaining_time":237.7254217,"test":[0.5223291914]}, +{"learn":[0.4355540706],"iteration":385,"passed_time":82.29253941,"remaining_time":237.4971215,"test":[0.5223007769]}, +{"learn":[0.4353642283],"iteration":386,"passed_time":82.50395683,"remaining_time":237.2788216,"test":[0.52229511]}, +{"learn":[0.4352438734],"iteration":387,"passed_time":82.70863509,"remaining_time":237.0412428,"test":[0.522239199]}, +{"learn":[0.4351800116],"iteration":388,"passed_time":82.92535806,"remaining_time":236.8382334,"test":[0.5223085989]}, +{"learn":[0.4350318194],"iteration":389,"passed_time":83.13880078,"remaining_time":236.6258176,"test":[0.5222770716]}, +{"learn":[0.4348216935],"iteration":390,"passed_time":83.37736594,"remaining_time":236.4846517,"test":[0.5223066035]}, +{"learn":[0.4347826317],"iteration":391,"passed_time":83.57727806,"remaining_time":236.2337349,"test":[0.5222492159]}, +{"learn":[0.4346708603],"iteration":392,"passed_time":83.78045675,"remaining_time":235.9922789,"test":[0.5222821]}, +{"learn":[0.434634598],"iteration":393,"passed_time":83.97666429,"remaining_time":235.7314485,"test":[0.5222827386]}, +{"learn":[0.4344988191],"iteration":394,"passed_time":84.18542104,"remaining_time":235.5060513,"test":[0.5222561998]}, +{"learn":[0.4343509175],"iteration":395,"passed_time":84.39099982,"remaining_time":235.2718783,"test":[0.5222660172]}, +{"learn":[0.4342753292],"iteration":396,"passed_time":84.60169294,"remaining_time":235.0520587,"test":[0.5222762735]}, +{"learn":[0.4341275068],"iteration":397,"passed_time":84.80834126,"remaining_time":234.8210856,"test":[0.5222696089]}, +{"learn":[0.4339127854],"iteration":398,"passed_time":85.01703362,"remaining_time":234.5958747,"test":[0.5221320862]}, +{"learn":[0.4338583523],"iteration":399,"passed_time":85.22540995,"remaining_time":234.3698774,"test":[0.5221256211]}, +{"learn":[0.4336283653],"iteration":400,"passed_time":85.43933326,"remaining_time":234.1591702,"test":[0.5221399082]}, +{"learn":[0.4335961703],"iteration":401,"passed_time":85.64723756,"remaining_time":233.9320071,"test":[0.5221409458]}, +{"learn":[0.4334266645],"iteration":402,"passed_time":85.90194129,"remaining_time":233.8323315,"test":[0.5221340018]}, +{"learn":[0.4333842749],"iteration":403,"passed_time":86.13668494,"remaining_time":233.6777393,"test":[0.5220711868]}, +{"learn":[0.4333006312],"iteration":404,"passed_time":86.35359131,"remaining_time":233.4745247,"test":[0.5221006388]}, +{"learn":[0.4332627314],"iteration":405,"passed_time":86.5552568,"remaining_time":233.2301747,"test":[0.5221415843]}, +{"learn":[0.433220738],"iteration":406,"passed_time":86.75534549,"remaining_time":232.9818,"test":[0.522124384]}, +{"learn":[0.433174281],"iteration":407,"passed_time":86.9551908,"remaining_time":232.7330107,"test":[0.5221449765]}, +{"learn":[0.4331175766],"iteration":408,"passed_time":87.15780313,"remaining_time":232.4918416,"test":[0.5221361169]}, +{"learn":[0.4330016059],"iteration":409,"passed_time":87.37740799,"remaining_time":232.2960359,"test":[0.5221247831]}, +{"learn":[0.4329391439],"iteration":410,"passed_time":87.57932833,"remaining_time":232.0532568,"test":[0.5221431407]}, +{"learn":[0.4327437025],"iteration":411,"passed_time":87.80340196,"remaining_time":231.869178,"test":[0.5221476104]}, +{"learn":[0.4326503132],"iteration":412,"passed_time":88.01088689,"remaining_time":231.6412447,"test":[0.522193704]}, +{"learn":[0.4326064973],"iteration":413,"passed_time":88.21140418,"remaining_time":231.3951327,"test":[0.5221899128]}, +{"learn":[0.4325309883],"iteration":414,"passed_time":88.41530448,"remaining_time":231.1580852,"test":[0.522223635]}, +{"learn":[0.4322932629],"iteration":415,"passed_time":88.63565927,"remaining_time":230.9640737,"test":[0.5222830578]}, +{"learn":[0.4321530469],"iteration":416,"passed_time":88.84510786,"remaining_time":230.7416111,"test":[0.5222495751]}, +{"learn":[0.4320995382],"iteration":417,"passed_time":89.0448387,"remaining_time":230.4940561,"test":[0.5222620663]}, +{"learn":[0.432064121],"iteration":418,"passed_time":89.24994153,"remaining_time":230.260589,"test":[0.5222605099]}, +{"learn":[0.4318960943],"iteration":419,"passed_time":89.47922216,"remaining_time":230.0894284,"test":[0.5222401967]}, +{"learn":[0.4317620056],"iteration":420,"passed_time":89.69071914,"remaining_time":229.8724132,"test":[0.5223113924]}, +{"learn":[0.4316429448],"iteration":421,"passed_time":89.93389099,"remaining_time":229.7363377,"test":[0.5222913986]}, +{"learn":[0.4316029322],"iteration":422,"passed_time":90.1689221,"remaining_time":229.5790286,"test":[0.5223001384]}, +{"learn":[0.4314649347],"iteration":423,"passed_time":90.37204545,"remaining_time":229.3403795,"test":[0.5222907202]}, +{"learn":[0.4313358113],"iteration":424,"passed_time":90.57688424,"remaining_time":229.1062366,"test":[0.5222583947]}, +{"learn":[0.4312889846],"iteration":425,"passed_time":90.78019257,"remaining_time":228.8683728,"test":[0.5222463426]}, +{"learn":[0.4311176301],"iteration":426,"passed_time":90.98763985,"remaining_time":228.6410716,"test":[0.5221930256]}, +{"learn":[0.4310804698],"iteration":427,"passed_time":91.19045567,"remaining_time":228.4022628,"test":[0.5221907908]}, +{"learn":[0.4309550439],"iteration":428,"passed_time":91.39238231,"remaining_time":228.161402,"test":[0.5221801353]}, +{"learn":[0.4308267393],"iteration":429,"passed_time":91.5943256,"remaining_time":227.9207637,"test":[0.5222163318]}, +{"learn":[0.4306758797],"iteration":430,"passed_time":91.80276972,"remaining_time":227.6964288,"test":[0.5222692497]}, +{"learn":[0.430564214],"iteration":431,"passed_time":92.01850188,"remaining_time":227.4901852,"test":[0.5222568782]}, +{"learn":[0.4305215074],"iteration":432,"passed_time":92.25172261,"remaining_time":227.3269931,"test":[0.5222648598]}, +{"learn":[0.4304761861],"iteration":433,"passed_time":92.4623496,"remaining_time":227.1079831,"test":[0.5222783088]}, +{"learn":[0.4303507603],"iteration":434,"passed_time":92.67243333,"remaining_time":226.8876816,"test":[0.5222841353]}, +{"learn":[0.4302473348],"iteration":435,"passed_time":92.88290673,"remaining_time":226.6683779,"test":[0.5222135383]}, +{"learn":[0.4302070316],"iteration":436,"passed_time":93.08297112,"remaining_time":226.4237947,"test":[0.5222171699]}, +{"learn":[0.4301724068],"iteration":437,"passed_time":93.29838769,"remaining_time":226.2166386,"test":[0.5222277056]}, +{"learn":[0.4300275424],"iteration":438,"passed_time":93.52747347,"remaining_time":226.0424814,"test":[0.5222634231]}, +{"learn":[0.4299943702],"iteration":439,"passed_time":93.72284133,"remaining_time":225.786845,"test":[0.5223040494]}, +{"learn":[0.4298901524],"iteration":440,"passed_time":93.93586475,"remaining_time":225.5738793,"test":[0.5222941522]}, +{"learn":[0.4298619455],"iteration":441,"passed_time":94.1548193,"remaining_time":225.3751105,"test":[0.5222503334]}, +{"learn":[0.4297966047],"iteration":442,"passed_time":94.35959653,"remaining_time":225.1424233,"test":[0.5222784285]}, +{"learn":[0.42959427],"iteration":443,"passed_time":94.57820777,"remaining_time":224.9427644,"test":[0.522256559]}, +{"learn":[0.4293884491],"iteration":444,"passed_time":94.78428171,"remaining_time":224.7132971,"test":[0.5222131392]}, +{"learn":[0.4292061075],"iteration":445,"passed_time":94.9988964,"remaining_time":224.5041184,"test":[0.5222200033]}, +{"learn":[0.4291471318],"iteration":446,"passed_time":95.20125446,"remaining_time":224.2660424,"test":[0.5222137777]}, +{"learn":[0.4290125149],"iteration":447,"passed_time":95.4125096,"remaining_time":224.0490181,"test":[0.5221952205]}, +{"learn":[0.4288033134],"iteration":448,"passed_time":95.63081023,"remaining_time":223.8485113,"test":[0.5222177286]}, +{"learn":[0.4287767175],"iteration":449,"passed_time":95.82847563,"remaining_time":223.5997765,"test":[0.5222333326]}, +{"learn":[0.4287522345],"iteration":450,"passed_time":96.02562344,"remaining_time":223.3500643,"test":[0.5222451054]}, +{"learn":[0.4287100297],"iteration":451,"passed_time":96.22047964,"remaining_time":223.0952714,"test":[0.5222368046]}, +{"learn":[0.4284893922],"iteration":452,"passed_time":96.42532174,"remaining_time":222.8638231,"test":[0.5221670855]}, +{"learn":[0.4284411129],"iteration":453,"passed_time":96.62269637,"remaining_time":222.6152872,"test":[0.5221496058]}, +{"learn":[0.4284061448],"iteration":454,"passed_time":96.82154444,"remaining_time":222.3703603,"test":[0.5221062658]}, +{"learn":[0.4282836505],"iteration":455,"passed_time":97.02182345,"remaining_time":222.1289116,"test":[0.5221362366]}, +{"learn":[0.42796347],"iteration":456,"passed_time":97.23581219,"remaining_time":221.9189324,"test":[0.5221121722]}, +{"learn":[0.427910648],"iteration":457,"passed_time":97.43251092,"remaining_time":221.6695991,"test":[0.5220806848]}, +{"learn":[0.4276986997],"iteration":458,"passed_time":97.64809897,"remaining_time":221.4633356,"test":[0.5220582167]}, +{"learn":[0.4275380416],"iteration":459,"passed_time":97.8527668,"remaining_time":221.2323423,"test":[0.5221086603]}, +{"learn":[0.4274078618],"iteration":460,"passed_time":98.05854056,"remaining_time":221.0039558,"test":[0.5220843165]}, +{"learn":[0.4272547308],"iteration":461,"passed_time":98.27073606,"remaining_time":220.7900953,"test":[0.5220927769]}, +{"learn":[0.4272262333],"iteration":462,"passed_time":98.49014829,"remaining_time":220.5924056,"test":[0.5220818422]}, +{"learn":[0.4271944609],"iteration":463,"passed_time":98.68969373,"remaining_time":220.3502644,"test":[0.5220434907]}, +{"learn":[0.4271767391],"iteration":464,"passed_time":98.73066173,"remaining_time":219.7553439,"test":[0.5220454462]}, +{"learn":[0.4271252641],"iteration":465,"passed_time":98.92478283,"remaining_time":219.5026297,"test":[0.5220438498]}, +{"learn":[0.4269907001],"iteration":466,"passed_time":99.12392397,"remaining_time":219.2612708,"test":[0.5220508337]}, +{"learn":[0.4269561016],"iteration":467,"passed_time":99.31939585,"remaining_time":219.0120011,"test":[0.522099242]}, +{"learn":[0.4267656783],"iteration":468,"passed_time":99.52760972,"remaining_time":218.7909715,"test":[0.5220674753]}, +{"learn":[0.4266574461],"iteration":469,"passed_time":99.73766708,"remaining_time":218.5740364,"test":[0.5220163932]}, +{"learn":[0.4266170901],"iteration":470,"passed_time":99.93231392,"remaining_time":218.3234629,"test":[0.5219874999]}, +{"learn":[0.4264592051],"iteration":471,"passed_time":100.1664133,"remaining_time":218.1590527,"test":[0.5220094492]}, +{"learn":[0.4263245354],"iteration":472,"passed_time":100.3701333,"remaining_time":217.9283867,"test":[0.5219871407]}, +{"learn":[0.4262025958],"iteration":473,"passed_time":100.5719726,"remaining_time":217.6937634,"test":[0.5219403287]}, +{"learn":[0.4261693444],"iteration":474,"passed_time":100.7727919,"remaining_time":217.4570772,"test":[0.521902935]}, +{"learn":[0.4259847315],"iteration":475,"passed_time":101.0086863,"remaining_time":217.2959974,"test":[0.5218629074]}, +{"learn":[0.4259156403],"iteration":476,"passed_time":101.2068856,"remaining_time":217.0537609,"test":[0.5218597946]}, +{"learn":[0.4258481866],"iteration":477,"passed_time":101.4148864,"remaining_time":216.832665,"test":[0.5218983456]}, +{"learn":[0.4258218284],"iteration":478,"passed_time":101.6256538,"remaining_time":216.6175209,"test":[0.5219061277]}, +{"learn":[0.4257969757],"iteration":479,"passed_time":101.8253851,"remaining_time":216.3789434,"test":[0.5218939558]}, +{"learn":[0.425748802],"iteration":480,"passed_time":102.0370173,"remaining_time":216.1657393,"test":[0.5219066465]}, +{"learn":[0.4256836989],"iteration":481,"passed_time":102.2371616,"remaining_time":215.928279,"test":[0.5219238867]}, +{"learn":[0.425660827],"iteration":482,"passed_time":102.4345382,"remaining_time":215.6851457,"test":[0.5218900049]}, +{"learn":[0.4255501385],"iteration":483,"passed_time":102.6812186,"remaining_time":215.5456986,"test":[0.521844869]}, +{"learn":[0.4255061906],"iteration":484,"passed_time":102.8815831,"remaining_time":215.3088801,"test":[0.5218519726]}, +{"learn":[0.4253049652],"iteration":485,"passed_time":103.0897672,"remaining_time":215.0885266,"test":[0.5218498575]}, +{"learn":[0.4251170245],"iteration":486,"passed_time":103.2989851,"remaining_time":214.8703735,"test":[0.5218545667]}, +{"learn":[0.4249721602],"iteration":487,"passed_time":103.511939,"remaining_time":214.6600046,"test":[0.5217666495]}, +{"learn":[0.4249461453],"iteration":488,"passed_time":103.7109515,"remaining_time":214.4208015,"test":[0.5217848875]}, +{"learn":[0.4249259937],"iteration":489,"passed_time":103.9145353,"remaining_time":214.1911851,"test":[0.5217765866]}, +{"learn":[0.4248146186],"iteration":490,"passed_time":104.1407356,"remaining_time":214.0081512,"test":[0.5217925099]}, +{"learn":[0.4247909279],"iteration":491,"passed_time":104.3752024,"remaining_time":213.8418781,"test":[0.5218065574]}, +{"learn":[0.4247375248],"iteration":492,"passed_time":104.5835651,"remaining_time":213.6220083,"test":[0.5217701215]}, +{"learn":[0.4246067904],"iteration":493,"passed_time":104.7929014,"remaining_time":213.4041677,"test":[0.5218236779]}, +{"learn":[0.424525286],"iteration":494,"passed_time":105.0091307,"remaining_time":213.2003563,"test":[0.5218390824]}, +{"learn":[0.4243306898],"iteration":495,"passed_time":105.2205624,"remaining_time":212.9867836,"test":[0.5217997731]}, +{"learn":[0.4242090143],"iteration":496,"passed_time":105.4284495,"remaining_time":212.7660661,"test":[0.5218265114]}, +{"learn":[0.4241103164],"iteration":497,"passed_time":105.6378255,"remaining_time":212.5483959,"test":[0.5218737624]}, +{"learn":[0.4240750049],"iteration":498,"passed_time":105.8365865,"remaining_time":212.3094651,"test":[0.5218910026]}, +{"learn":[0.423876315],"iteration":499,"passed_time":106.0471694,"remaining_time":212.0943388,"test":[0.5218790302]}, +{"learn":[0.4236666909],"iteration":500,"passed_time":106.2560012,"remaining_time":211.8757389,"test":[0.5218104684]}, +{"learn":[0.4235874314],"iteration":501,"passed_time":106.4587083,"remaining_time":211.6450017,"test":[0.5218612313]}, +{"learn":[0.4235354546],"iteration":502,"passed_time":106.6586536,"remaining_time":211.4089018,"test":[0.5218548859]}, +{"learn":[0.4234143338],"iteration":503,"passed_time":106.9082809,"remaining_time":211.2711266,"test":[0.5218411177]}, +{"learn":[0.4233029057],"iteration":504,"passed_time":107.114492,"remaining_time":211.0473654,"test":[0.5218926787]}, +{"learn":[0.423176714],"iteration":505,"passed_time":107.331729,"remaining_time":210.8453333,"test":[0.5218558836]}, +{"learn":[0.4231508576],"iteration":506,"passed_time":107.5341524,"remaining_time":210.6142274,"test":[0.5218650624]}, +{"learn":[0.4229899882],"iteration":507,"passed_time":107.7367559,"remaining_time":210.3835863,"test":[0.5218238775]}, +{"learn":[0.4228718518],"iteration":508,"passed_time":107.9435918,"remaining_time":210.1612957,"test":[0.5218550455]}, +{"learn":[0.4228359593],"iteration":509,"passed_time":108.1429389,"remaining_time":209.9245285,"test":[0.521825394]}, +{"learn":[0.4227992216],"iteration":510,"passed_time":108.345077,"remaining_time":209.6933095,"test":[0.5218180908]}, +{"learn":[0.422772388],"iteration":511,"passed_time":108.5467101,"remaining_time":209.4612296,"test":[0.5218546066]}, +{"learn":[0.4225437216],"iteration":512,"passed_time":108.7780629,"remaining_time":209.2864485,"test":[0.521883021]}, +{"learn":[0.4224996152],"iteration":513,"passed_time":108.9772527,"remaining_time":209.0497492,"test":[0.5218839389]}, +{"learn":[0.4223835653],"iteration":514,"passed_time":109.1913893,"remaining_time":208.8417834,"test":[0.5218613909]}, +{"learn":[0.4223577089],"iteration":515,"passed_time":109.3947618,"remaining_time":208.6132668,"test":[0.5218823825]}, +{"learn":[0.422249371],"iteration":516,"passed_time":109.5981148,"remaining_time":208.3848102,"test":[0.5218802274]}, +{"learn":[0.4222281101],"iteration":517,"passed_time":109.7971851,"remaining_time":208.1483317,"test":[0.5219119542]}, +{"learn":[0.422172198],"iteration":518,"passed_time":110.0078385,"remaining_time":207.9338913,"test":[0.5218955521]}, +{"learn":[0.4221387617],"iteration":519,"passed_time":110.2361452,"remaining_time":207.7527353,"test":[0.5219027754]}, +{"learn":[0.4221057744],"iteration":520,"passed_time":110.4566047,"remaining_time":207.5566525,"test":[0.521904691]}, +{"learn":[0.4219720291],"iteration":521,"passed_time":110.6619516,"remaining_time":207.3321621,"test":[0.5218664193]}, +{"learn":[0.4218148308],"iteration":522,"passed_time":110.8981992,"remaining_time":207.1654696,"test":[0.5218799082]}, +{"learn":[0.4216788934],"iteration":523,"passed_time":111.1573257,"remaining_time":207.0411256,"test":[0.5218879297]}, +{"learn":[0.4215360627],"iteration":524,"passed_time":111.3926584,"remaining_time":206.8720799,"test":[0.5218920003]}, +{"learn":[0.4214808109],"iteration":525,"passed_time":111.5949525,"remaining_time":206.6416041,"test":[0.5219174216]}, +{"learn":[0.4213864444],"iteration":526,"passed_time":111.8049091,"remaining_time":206.4253825,"test":[0.5219216518]}, +{"learn":[0.4212872711],"iteration":527,"passed_time":112.0050764,"remaining_time":206.1911634,"test":[0.521982272]}, +{"learn":[0.4212450927],"iteration":528,"passed_time":112.2035998,"remaining_time":205.9540555,"test":[0.522031638]}, +{"learn":[0.421218312],"iteration":529,"passed_time":112.4018752,"remaining_time":205.7166396,"test":[0.5220599726]}, +{"learn":[0.421077647],"iteration":530,"passed_time":112.6081136,"remaining_time":205.4939022,"test":[0.5220625267]}, +{"learn":[0.420989487],"iteration":531,"passed_time":112.8263081,"remaining_time":205.2929816,"test":[0.5220547447]}, +{"learn":[0.4209177812],"iteration":532,"passed_time":113.0358716,"remaining_time":205.0763374,"test":[0.5220632451]}, +{"learn":[0.4208866426],"iteration":533,"passed_time":113.2830326,"remaining_time":204.9277331,"test":[0.5220848752]}, +{"learn":[0.4208589903],"iteration":534,"passed_time":113.4939708,"remaining_time":204.7134239,"test":[0.5220681937]}, +{"learn":[0.42072197],"iteration":535,"passed_time":113.7005694,"remaining_time":204.4913227,"test":[0.5220367462]}, +{"learn":[0.4205736194],"iteration":536,"passed_time":113.9185212,"remaining_time":204.2896385,"test":[0.522055463]}, +{"learn":[0.4205432204],"iteration":537,"passed_time":114.1196441,"remaining_time":204.0578022,"test":[0.5220508736]}, +{"learn":[0.4205065355],"iteration":538,"passed_time":114.3166811,"remaining_time":203.818795,"test":[0.5220928168]}, +{"learn":[0.4204168966],"iteration":539,"passed_time":114.520698,"remaining_time":203.5923521,"test":[0.5221466526]}, +{"learn":[0.4203891914],"iteration":540,"passed_time":114.7212599,"remaining_time":203.3598674,"test":[0.5221287739]}, +{"learn":[0.420270923],"iteration":541,"passed_time":114.9328359,"remaining_time":203.1469683,"test":[0.5221223487]}, +{"learn":[0.4202239642],"iteration":542,"passed_time":115.1586282,"remaining_time":202.9591293,"test":[0.5221387908]}, +{"learn":[0.4201979758],"iteration":543,"passed_time":115.3764563,"remaining_time":202.7571547,"test":[0.5221799358]}, +{"learn":[0.4201635359],"iteration":544,"passed_time":115.5755957,"remaining_time":202.5223741,"test":[0.5221458145]}, +{"learn":[0.4201325293],"iteration":545,"passed_time":115.7756325,"remaining_time":202.289292,"test":[0.5221054677]}, +{"learn":[0.4201024736],"iteration":546,"passed_time":115.9806301,"remaining_time":202.0649735,"test":[0.5221112543]}, +{"learn":[0.4199850503],"iteration":547,"passed_time":116.2210864,"remaining_time":201.9023252,"test":[0.5220737808]}, +{"learn":[0.4198871183],"iteration":548,"passed_time":116.4555684,"remaining_time":201.7290446,"test":[0.5220705881]}, +{"learn":[0.4197798104],"iteration":549,"passed_time":116.6649609,"remaining_time":201.5122052,"test":[0.5220880678]}, +{"learn":[0.4196252004],"iteration":550,"passed_time":116.9190949,"remaining_time":201.372452,"test":[0.5220916196]}, +{"learn":[0.4195157797],"iteration":551,"passed_time":117.1264185,"remaining_time":201.1518926,"test":[0.522113569]}, +{"learn":[0.4193934175],"iteration":552,"passed_time":117.3342698,"remaining_time":200.9322848,"test":[0.5220915797]}, +{"learn":[0.4192236476],"iteration":553,"passed_time":117.5641319,"remaining_time":200.7503046,"test":[0.5220772129]}, +{"learn":[0.4191110311],"iteration":554,"passed_time":117.7710317,"remaining_time":200.529054,"test":[0.5221100571]}, +{"learn":[0.419012386],"iteration":555,"passed_time":118.0198404,"remaining_time":200.3790096,"test":[0.5220726634]}, +{"learn":[0.4189145333],"iteration":556,"passed_time":118.2231725,"remaining_time":200.1516188,"test":[0.5220840371]}, +{"learn":[0.4188031317],"iteration":557,"passed_time":118.4285758,"remaining_time":199.9278108,"test":[0.5220930164]}, +{"learn":[0.4185741747],"iteration":558,"passed_time":118.6485077,"remaining_time":199.7285255,"test":[0.5220205436]}, +{"learn":[0.4185236241],"iteration":559,"passed_time":118.856107,"remaining_time":199.5084654,"test":[0.5220209826]}, +{"learn":[0.4184201458],"iteration":560,"passed_time":119.0713845,"remaining_time":199.3013014,"test":[0.522004341]}, +{"learn":[0.4183944214],"iteration":561,"passed_time":119.3114278,"remaining_time":199.1354435,"test":[0.5220604515]}, +{"learn":[0.4183447159],"iteration":562,"passed_time":119.5773688,"remaining_time":199.0124238,"test":[0.5220947723]}, +{"learn":[0.4182208219],"iteration":563,"passed_time":119.784044,"remaining_time":198.7905411,"test":[0.5220045007]}, +{"learn":[0.4180463773],"iteration":564,"passed_time":119.9997259,"remaining_time":198.5836173,"test":[0.5220480402]}, +{"learn":[0.4178770828],"iteration":565,"passed_time":120.2305546,"remaining_time":198.4016572,"test":[0.5220247739]}, +{"learn":[0.4177670017],"iteration":566,"passed_time":120.434918,"remaining_time":198.1759762,"test":[0.5220191868]}, +{"learn":[0.4177384779],"iteration":567,"passed_time":120.6392877,"remaining_time":197.9503805,"test":[0.5220286848]}, +{"learn":[0.4175663046],"iteration":568,"passed_time":120.8471325,"remaining_time":197.7305455,"test":[0.5220263702]}, +{"learn":[0.4174468476],"iteration":569,"passed_time":121.0550746,"remaining_time":197.5109111,"test":[0.5220041016]}, +{"learn":[0.41726823],"iteration":570,"passed_time":121.2697685,"remaining_time":197.3023028,"test":[0.5220519112]}, +{"learn":[0.4171849561],"iteration":571,"passed_time":121.472815,"remaining_time":197.0747767,"test":[0.5221098974]}, +{"learn":[0.417080395],"iteration":572,"passed_time":121.6767546,"remaining_time":196.848781,"test":[0.5221239849]}, +{"learn":[0.4170362886],"iteration":573,"passed_time":121.8794938,"remaining_time":196.6209255,"test":[0.5221256611]}, +{"learn":[0.416929641],"iteration":574,"passed_time":122.0924276,"remaining_time":196.4095574,"test":[0.5221025145]}, +{"learn":[0.4168813353],"iteration":575,"passed_time":122.3158374,"remaining_time":196.2149891,"test":[0.522133922]}, +{"learn":[0.416801442],"iteration":576,"passed_time":122.5269062,"remaining_time":196.0005796,"test":[0.5221316472]}, +{"learn":[0.4166510049],"iteration":577,"passed_time":122.7275117,"remaining_time":195.769491,"test":[0.5221554324]}, +{"learn":[0.4165655125],"iteration":578,"passed_time":122.9414125,"remaining_time":195.5596561,"test":[0.5222276657]}, +{"learn":[0.4164508624],"iteration":579,"passed_time":123.1449403,"remaining_time":195.3333535,"test":[0.5222340509]}, +{"learn":[0.4163564431],"iteration":580,"passed_time":123.3717894,"remaining_time":195.144018,"test":[0.5222625851]}, +{"learn":[0.4163250932],"iteration":581,"passed_time":123.6043487,"remaining_time":194.9635603,"test":[0.5222746772]}, +{"learn":[0.4161199854],"iteration":582,"passed_time":123.8446786,"remaining_time":194.7951463,"test":[0.522315503]}, +{"learn":[0.4160572857],"iteration":583,"passed_time":124.04866,"remaining_time":194.5694736,"test":[0.5223301492]}, +{"learn":[0.4160180125],"iteration":584,"passed_time":124.2505874,"remaining_time":194.3406624,"test":[0.5223704961]}, +{"learn":[0.4159334445],"iteration":585,"passed_time":124.4521767,"remaining_time":194.1114155,"test":[0.5223333019]}, +{"learn":[0.4157927795],"iteration":586,"passed_time":124.6722736,"remaining_time":193.9110491,"test":[0.5223089182]}, +{"learn":[0.41569321],"iteration":587,"passed_time":124.8821689,"remaining_time":193.6947926,"test":[0.5222736396]}, +{"learn":[0.4154710407],"iteration":588,"passed_time":125.1048793,"remaining_time":193.4983788,"test":[0.5223176181]}, +{"learn":[0.4154438638],"iteration":589,"passed_time":125.3307553,"remaining_time":193.3067582,"test":[0.5223279542]}, +{"learn":[0.4153113598],"iteration":590,"passed_time":125.5395494,"remaining_time":193.0887486,"test":[0.5223896119]}, +{"learn":[0.4152067194],"iteration":591,"passed_time":125.745842,"remaining_time":192.8669333,"test":[0.5223692988]}, +{"learn":[0.4151275657],"iteration":592,"passed_time":125.9531098,"remaining_time":192.6466621,"test":[0.5223694984]}, +{"learn":[0.4149858442],"iteration":593,"passed_time":126.1591558,"remaining_time":192.424571,"test":[0.5224234538]}, +{"learn":[0.4149130819],"iteration":594,"passed_time":126.3637253,"remaining_time":192.200288,"test":[0.5224167493]}, +{"learn":[0.4146729531],"iteration":595,"passed_time":126.5860955,"remaining_time":192.003071,"test":[0.5224341891]}, +{"learn":[0.4144589976],"iteration":596,"passed_time":126.7955773,"remaining_time":191.7862753,"test":[0.5223873771]}, +{"learn":[0.4143572096],"iteration":597,"passed_time":127.0069224,"remaining_time":191.5723144,"test":[0.5224070916]}, +{"learn":[0.4143302968],"iteration":598,"passed_time":127.2067859,"remaining_time":191.341092,"test":[0.522377879]}, +{"learn":[0.4142094665],"iteration":599,"passed_time":127.4130411,"remaining_time":191.1195616,"test":[0.5223658268]}, +{"learn":[0.4140924393],"iteration":600,"passed_time":127.6190694,"remaining_time":190.8977427,"test":[0.5223324638]}, +{"learn":[0.4139592487],"iteration":601,"passed_time":127.8719477,"remaining_time":190.7458621,"test":[0.5223589627]}, +{"learn":[0.4138737299],"iteration":602,"passed_time":128.095432,"remaining_time":190.5499212,"test":[0.5223782781]}, +{"learn":[0.4138459455],"iteration":603,"passed_time":128.2992101,"remaining_time":190.324656,"test":[0.5223461123]}, +{"learn":[0.4136362157],"iteration":604,"passed_time":128.5311791,"remaining_time":190.1411658,"test":[0.5223435582]}, +{"learn":[0.4136080352],"iteration":605,"passed_time":128.7332051,"remaining_time":189.9133421,"test":[0.5223463518]}, +{"learn":[0.413524418],"iteration":606,"passed_time":128.9358972,"remaining_time":189.6865835,"test":[0.5223482274]}, +{"learn":[0.4133501054],"iteration":607,"passed_time":129.1424738,"remaining_time":189.465603,"test":[0.5223358959]}, +{"learn":[0.4132487135],"iteration":608,"passed_time":129.3465764,"remaining_time":189.2410502,"test":[0.5223318253]}, +{"learn":[0.4131160247],"iteration":609,"passed_time":129.6068737,"remaining_time":189.0985534,"test":[0.5223568476]}, +{"learn":[0.4130051249],"iteration":610,"passed_time":129.8167486,"remaining_time":188.8823069,"test":[0.5223686204]}, +{"learn":[0.4129740127],"iteration":611,"passed_time":130.0166065,"remaining_time":188.6515466,"test":[0.5223942413]}, +{"learn":[0.4128680782],"iteration":612,"passed_time":130.2455772,"remaining_time":188.463013,"test":[0.5223829074]}, +{"learn":[0.4128373622],"iteration":613,"passed_time":130.4566745,"remaining_time":188.2485564,"test":[0.5224129182]}, +{"learn":[0.4127119892],"iteration":614,"passed_time":130.6659731,"remaining_time":188.0315222,"test":[0.5223947601]}, +{"learn":[0.4126791075],"iteration":615,"passed_time":130.8693445,"remaining_time":187.8060073,"test":[0.5224285621]}, +{"learn":[0.4125751274],"iteration":616,"passed_time":131.0986367,"remaining_time":187.6176599,"test":[0.522424691]}, +{"learn":[0.4124947587],"iteration":617,"passed_time":131.3047381,"remaining_time":187.3960825,"test":[0.52240777]}, +{"learn":[0.4123330442],"iteration":618,"passed_time":131.5738127,"remaining_time":187.2641825,"test":[0.5224195828]}, +{"learn":[0.4123064483],"iteration":619,"passed_time":131.7719512,"remaining_time":187.0311565,"test":[0.522398671]}, +{"learn":[0.41213755],"iteration":620,"passed_time":132.0068018,"remaining_time":186.8502074,"test":[0.5224034201]}, +{"learn":[0.4120677985],"iteration":621,"passed_time":132.2355948,"remaining_time":186.6605341,"test":[0.5223694584]}, +{"learn":[0.4119360076],"iteration":622,"passed_time":132.4473759,"remaining_time":186.4467876,"test":[0.5223874569]}, +{"learn":[0.4118715648],"iteration":623,"passed_time":132.6533262,"remaining_time":186.2248617,"test":[0.5223998284]}, +{"learn":[0.4117428904],"iteration":624,"passed_time":132.8613342,"remaining_time":186.0058679,"test":[0.5223841845]}, +{"learn":[0.411610941],"iteration":625,"passed_time":133.0772821,"remaining_time":185.7979945,"test":[0.5223773203]}, +{"learn":[0.4115017844],"iteration":626,"passed_time":133.2912884,"remaining_time":185.587392,"test":[0.5223327831]}, +{"learn":[0.411481131],"iteration":627,"passed_time":133.4887607,"remaining_time":185.3538207,"test":[0.5223507815]}, +{"learn":[0.4113286075],"iteration":628,"passed_time":133.7001987,"remaining_time":185.1397028,"test":[0.5223174584]}, +{"learn":[0.4112462844],"iteration":629,"passed_time":133.9191715,"remaining_time":184.9359987,"test":[0.5223235244]}, +{"learn":[0.4110174331],"iteration":630,"passed_time":134.1376259,"remaining_time":184.7315323,"test":[0.5222586741]}, +{"learn":[0.4108394229],"iteration":631,"passed_time":134.3846641,"remaining_time":184.5662792,"test":[0.5222450655]}, +{"learn":[0.4105907898],"iteration":632,"passed_time":134.615146,"remaining_time":184.3780909,"test":[0.5222296211]}, +{"learn":[0.4105138809],"iteration":633,"passed_time":134.8265737,"remaining_time":184.1637426,"test":[0.5222178084]}, +{"learn":[0.4101628787],"iteration":634,"passed_time":135.0530537,"remaining_time":183.9699077,"test":[0.5222282643]}, +{"learn":[0.4099986023],"iteration":635,"passed_time":135.2659233,"remaining_time":183.7574808,"test":[0.522237483]}, +{"learn":[0.4098852727],"iteration":636,"passed_time":135.4733028,"remaining_time":183.5376144,"test":[0.5222393188]}, +{"learn":[0.4097632539],"iteration":637,"passed_time":135.6800792,"remaining_time":183.3169722,"test":[0.5222374032]}, +{"learn":[0.4096657445],"iteration":638,"passed_time":135.8888978,"remaining_time":183.0991252,"test":[0.5222169703]}, +{"learn":[0.4095621077],"iteration":639,"passed_time":136.1103566,"remaining_time":182.8982916,"test":[0.5222458238]}, +{"learn":[0.4094162134],"iteration":640,"passed_time":136.341365,"remaining_time":182.7101911,"test":[0.5221768629]}, +{"learn":[0.4093790531],"iteration":641,"passed_time":136.5422121,"remaining_time":182.481648,"test":[0.5221749473]}, +{"learn":[0.4092723527],"iteration":642,"passed_time":136.7478475,"remaining_time":182.2595728,"test":[0.5221539957]}, +{"learn":[0.4091630375],"iteration":643,"passed_time":136.9529131,"remaining_time":182.0367914,"test":[0.5221308491]}, +{"learn":[0.4090574464],"iteration":644,"passed_time":137.1736507,"remaining_time":181.8348392,"test":[0.5220680739]}, +{"learn":[0.4089596728],"iteration":645,"passed_time":137.4034809,"remaining_time":181.6448494,"test":[0.5220853541]}, +{"learn":[0.4088230751],"iteration":646,"passed_time":137.6166572,"remaining_time":181.4327799,"test":[0.5221375536]}, +{"learn":[0.4088042177],"iteration":647,"passed_time":137.8141338,"remaining_time":181.2000648,"test":[0.5221546741]}, +{"learn":[0.4086547578],"iteration":648,"passed_time":138.0356342,"remaining_time":180.9989595,"test":[0.5221430609]}, +{"learn":[0.4084876819],"iteration":649,"passed_time":138.2393867,"remaining_time":180.7745826,"test":[0.5221245835]}, +{"learn":[0.4084118559],"iteration":650,"passed_time":138.4439369,"remaining_time":180.5513094,"test":[0.5221498852]}, +{"learn":[0.4082863772],"iteration":651,"passed_time":138.6595013,"remaining_time":180.3424189,"test":[0.5221640525]}, +{"learn":[0.4082518316],"iteration":652,"passed_time":138.8592601,"remaining_time":180.1130066,"test":[0.5221665267]}, +{"learn":[0.4080320657],"iteration":653,"passed_time":139.0755687,"remaining_time":179.9050935,"test":[0.5221492865]}, +{"learn":[0.4078377072],"iteration":654,"passed_time":139.2849104,"remaining_time":179.6881669,"test":[0.5221784991]}, +{"learn":[0.4076549166],"iteration":655,"passed_time":139.4985962,"remaining_time":179.4768524,"test":[0.5222281046]}, +{"learn":[0.4075353804],"iteration":656,"passed_time":139.7121479,"remaining_time":179.2653587,"test":[0.5222383211]}, +{"learn":[0.4074434437],"iteration":657,"passed_time":139.9272668,"remaining_time":179.0558642,"test":[0.5222958284]}, +{"learn":[0.4072420335],"iteration":658,"passed_time":140.1402871,"remaining_time":178.8436745,"test":[0.5223166204]}, +{"learn":[0.4071416188],"iteration":659,"passed_time":140.3556632,"remaining_time":178.6344805,"test":[0.5223578053]}, +{"learn":[0.407027312],"iteration":660,"passed_time":140.5796933,"remaining_time":178.4362522,"test":[0.5223855812]}, +{"learn":[0.4069112357],"iteration":661,"passed_time":140.7919198,"remaining_time":178.2230042,"test":[0.5223741676]}, +{"learn":[0.406769752],"iteration":662,"passed_time":141.0007769,"remaining_time":178.0055057,"test":[0.5223762827]}, +{"learn":[0.4065997972],"iteration":663,"passed_time":141.2094123,"remaining_time":177.787754,"test":[0.5223555705]}, +{"learn":[0.4064893464],"iteration":664,"passed_time":141.414348,"remaining_time":177.5653844,"test":[0.5223429596]}, +{"learn":[0.4062135099],"iteration":665,"passed_time":141.6308512,"remaining_time":177.3575524,"test":[0.5222659772]}, +{"learn":[0.4061617972],"iteration":666,"passed_time":141.8329938,"remaining_time":177.1317599,"test":[0.5222826188]}, +{"learn":[0.4060348659],"iteration":667,"passed_time":142.040605,"remaining_time":176.9128493,"test":[0.5222708061]}, +{"learn":[0.4060129448],"iteration":668,"passed_time":142.2397015,"remaining_time":176.683396,"test":[0.5222718038]}, +{"learn":[0.4059170464],"iteration":669,"passed_time":142.4513633,"remaining_time":176.4695994,"test":[0.5222055966]}, +{"learn":[0.4058050374],"iteration":670,"passed_time":142.6606606,"remaining_time":176.2528877,"test":[0.5222149749]}, +{"learn":[0.4056596712],"iteration":671,"passed_time":142.8727906,"remaining_time":176.0396884,"test":[0.5221594231]}, +{"learn":[0.4055701908],"iteration":672,"passed_time":143.0841098,"remaining_time":175.825496,"test":[0.5221519205]}, +{"learn":[0.4054526882],"iteration":673,"passed_time":143.2967632,"remaining_time":175.6129471,"test":[0.5221678038]}, +{"learn":[0.4053703916],"iteration":674,"passed_time":143.5166527,"remaining_time":175.4092422,"test":[0.5221557915]}, +{"learn":[0.4052619744],"iteration":675,"passed_time":143.7279505,"remaining_time":175.1950166,"test":[0.5221788184]}, +{"learn":[0.4051707244],"iteration":676,"passed_time":143.934227,"remaining_time":174.9746954,"test":[0.5222169304]}, +{"learn":[0.4050839114],"iteration":677,"passed_time":144.1434248,"remaining_time":174.7579575,"test":[0.5222110241]}, +{"learn":[0.4050310102],"iteration":678,"passed_time":144.3452055,"remaining_time":174.5322735,"test":[0.5221965774]}, +{"learn":[0.4049275319],"iteration":679,"passed_time":144.5622615,"remaining_time":174.32508,"test":[0.5221970164]}, +{"learn":[0.4048024494],"iteration":680,"passed_time":144.7898207,"remaining_time":174.1304892,"test":[0.5221466127]}, +{"learn":[0.4047124671],"iteration":681,"passed_time":144.9904908,"remaining_time":173.9035506,"test":[0.5221535966]}, +{"learn":[0.4045690818],"iteration":682,"passed_time":145.1971941,"remaining_time":173.6839056,"test":[0.5221871591]}, +{"learn":[0.4044315598],"iteration":683,"passed_time":145.4058108,"remaining_time":173.4665813,"test":[0.5221479297]}, +{"learn":[0.4042882801],"iteration":684,"passed_time":145.6149483,"remaining_time":173.249902,"test":[0.5220778514]}, +{"learn":[0.4041545084],"iteration":685,"passed_time":145.8277972,"remaining_time":173.0376486,"test":[0.5220880678]}, +{"learn":[0.4039426129],"iteration":686,"passed_time":146.0465538,"remaining_time":172.8323846,"test":[0.522129572]}, +{"learn":[0.4038427265],"iteration":687,"passed_time":146.2713765,"remaining_time":172.6342409,"test":[0.522132605]}, +{"learn":[0.4037263332],"iteration":688,"passed_time":146.5033202,"remaining_time":172.4444016,"test":[0.5221517209]}, +{"learn":[0.4036760467],"iteration":689,"passed_time":146.7078971,"remaining_time":172.222314,"test":[0.5221073034]}, +{"learn":[0.403583212],"iteration":690,"passed_time":146.9144884,"remaining_time":172.0026355,"test":[0.5221417439]}, +{"learn":[0.4034757985],"iteration":691,"passed_time":147.1178148,"remaining_time":171.7791826,"test":[0.5221258207]}, +{"learn":[0.4033590355],"iteration":692,"passed_time":147.3289845,"remaining_time":171.5649214,"test":[0.5221208721]}, +{"learn":[0.4032205626],"iteration":693,"passed_time":147.5317754,"remaining_time":171.3409379,"test":[0.5220777317]}, +{"learn":[0.4030926541],"iteration":694,"passed_time":147.7355699,"remaining_time":171.1181781,"test":[0.5221267386]}, +{"learn":[0.4029039211],"iteration":695,"passed_time":147.9445829,"remaining_time":170.901501,"test":[0.5221320064]}, +{"learn":[0.4028226545],"iteration":696,"passed_time":148.1591739,"remaining_time":170.6912721,"test":[0.5221535168]}, +{"learn":[0.4027551215],"iteration":697,"passed_time":148.358769,"remaining_time":170.4638005,"test":[0.5221580264]}, +{"learn":[0.4025954142],"iteration":698,"passed_time":148.5695473,"remaining_time":170.2492237,"test":[0.5221366756]}, +{"learn":[0.4024508404],"iteration":699,"passed_time":148.8168676,"remaining_time":170.0764201,"test":[0.5221112543]}, +{"learn":[0.4023101754],"iteration":700,"passed_time":149.0598128,"remaining_time":169.8984172,"test":[0.5220847954]}, +{"learn":[0.4022780332],"iteration":701,"passed_time":149.2652557,"remaining_time":169.6775984,"test":[0.5220865114]}, +{"learn":[0.4021717026],"iteration":702,"passed_time":149.473086,"remaining_time":169.45953,"test":[0.5220447677]}, +{"learn":[0.4020247253],"iteration":703,"passed_time":149.6904759,"remaining_time":169.2522995,"test":[0.5220336733]}, +{"learn":[0.4019478165],"iteration":704,"passed_time":149.8983718,"remaining_time":169.0343342,"test":[0.5220583364]}, +{"learn":[0.4017364757],"iteration":705,"passed_time":150.1146841,"remaining_time":168.8258628,"test":[0.5220547048]}, +{"learn":[0.4016885926],"iteration":706,"passed_time":150.3149406,"remaining_time":168.5993605,"test":[0.5220721845]}, +{"learn":[0.4015977123],"iteration":707,"passed_time":150.5235323,"remaining_time":168.3822565,"test":[0.5220881077]}, +{"learn":[0.4015706146],"iteration":708,"passed_time":150.7246451,"remaining_time":168.1568326,"test":[0.5221016365]}, +{"learn":[0.4015012329],"iteration":709,"passed_time":150.9328258,"remaining_time":167.9393413,"test":[0.522074978]}, +{"learn":[0.4013771805],"iteration":710,"passed_time":151.1376807,"remaining_time":167.7181857,"test":[0.5220769335]}, +{"learn":[0.4013263657],"iteration":711,"passed_time":151.3489088,"remaining_time":167.5041294,"test":[0.5220694308]}, +{"learn":[0.4012401338],"iteration":712,"passed_time":151.5539644,"remaining_time":167.2832678,"test":[0.5220553433]}, +{"learn":[0.401160082],"iteration":713,"passed_time":151.7593149,"remaining_time":167.0627752,"test":[0.5220847954]}, +{"learn":[0.4010814301],"iteration":714,"passed_time":151.9634811,"remaining_time":166.8410247,"test":[0.5222718038]}, +{"learn":[0.4009449116],"iteration":715,"passed_time":152.1734392,"remaining_time":166.6256653,"test":[0.5222875674]}, +{"learn":[0.4008109022],"iteration":716,"passed_time":152.4043753,"remaining_time":166.4332299,"test":[0.5222551223]}, +{"learn":[0.4006492669],"iteration":717,"passed_time":152.6176557,"remaining_time":166.2214579,"test":[0.5221990118]}, +{"learn":[0.400531421],"iteration":718,"passed_time":152.8210207,"remaining_time":165.9989112,"test":[0.522248737]}, +{"learn":[0.4003784485],"iteration":719,"passed_time":153.1321273,"remaining_time":165.8931379,"test":[0.5222378821]}, +{"learn":[0.4002264268],"iteration":720,"passed_time":153.3505969,"remaining_time":165.6867059,"test":[0.5222573172]}, +{"learn":[0.4001478805],"iteration":721,"passed_time":153.553079,"remaining_time":165.4630131,"test":[0.5222718836]}, +{"learn":[0.4000405726],"iteration":722,"passed_time":153.7845849,"remaining_time":165.2705705,"test":[0.5222899619]}, +{"learn":[0.3999716927],"iteration":723,"passed_time":153.9875382,"remaining_time":165.0474167,"test":[0.5223036503]}, +{"learn":[0.3998335896],"iteration":724,"passed_time":154.1958481,"remaining_time":164.8300446,"test":[0.5223023334]}, +{"learn":[0.3997517947],"iteration":725,"passed_time":154.4052791,"remaining_time":164.6138926,"test":[0.5222797056]}, +{"learn":[0.3997031721],"iteration":726,"passed_time":154.6040484,"remaining_time":164.3864229,"test":[0.5222763533]}, +{"learn":[0.3994946572],"iteration":727,"passed_time":154.817593,"remaining_time":164.1747002,"test":[0.5222781891]}, +{"learn":[0.3994120964],"iteration":728,"passed_time":155.0291318,"remaining_time":163.9608513,"test":[0.5222687708]}, +{"learn":[0.3992452317],"iteration":729,"passed_time":155.253666,"remaining_time":163.7607162,"test":[0.5222255106]}, +{"learn":[0.3991151311],"iteration":730,"passed_time":155.5078555,"remaining_time":163.5917111,"test":[0.5222417931]}, +{"learn":[0.398916943],"iteration":731,"passed_time":155.7293318,"remaining_time":163.3881514,"test":[0.5222338514]}, +{"learn":[0.3988042209],"iteration":732,"passed_time":155.9350046,"remaining_time":163.1680062,"test":[0.5221923871]}, +{"learn":[0.3987857067],"iteration":733,"passed_time":156.1348275,"remaining_time":162.9417955,"test":[0.5222242735]}, +{"learn":[0.398664269],"iteration":734,"passed_time":156.3439685,"remaining_time":162.7253549,"test":[0.5222261891]}, +{"learn":[0.3985673142],"iteration":735,"passed_time":156.5481671,"remaining_time":162.5038039,"test":[0.5221916288]}, +{"learn":[0.3984358931],"iteration":736,"passed_time":156.7558049,"remaining_time":162.2858604,"test":[0.5222121415]}, +{"learn":[0.3983485783],"iteration":737,"passed_time":156.9671,"remaining_time":162.0717212,"test":[0.5222113832]}, +{"learn":[0.3983240953],"iteration":738,"passed_time":157.1876098,"remaining_time":161.8670785,"test":[0.5222260294]}, +{"learn":[0.3982504614],"iteration":739,"passed_time":157.4191209,"remaining_time":161.6736918,"test":[0.5222251515]}, +{"learn":[0.3981624864],"iteration":740,"passed_time":157.6226995,"remaining_time":161.451591,"test":[0.5222304592]}, +{"learn":[0.3980996546],"iteration":741,"passed_time":157.8319623,"remaining_time":161.235347,"test":[0.5222567186]}, +{"learn":[0.3980128945],"iteration":742,"passed_time":158.0469054,"remaining_time":161.024909,"test":[0.5222550824]}, +{"learn":[0.3979900754],"iteration":743,"passed_time":158.2554619,"remaining_time":160.8079693,"test":[0.522232295]}, +{"learn":[0.3978655475],"iteration":744,"passed_time":158.4845851,"remaining_time":160.611895,"test":[0.5222211208]}, +{"learn":[0.3977282631],"iteration":745,"passed_time":158.70236,"remaining_time":160.404262,"test":[0.5222276258]}, +{"learn":[0.3976407107],"iteration":746,"passed_time":158.9214057,"remaining_time":160.1978829,"test":[0.5222530471]}, +{"learn":[0.3974671112],"iteration":747,"passed_time":159.1317647,"remaining_time":159.9827367,"test":[0.5223028522]}, +{"learn":[0.3973722429],"iteration":748,"passed_time":159.3334345,"remaining_time":159.7588909,"test":[0.5223244423]}, +{"learn":[0.3972935381],"iteration":749,"passed_time":159.5395776,"remaining_time":159.5395776,"test":[0.5223265175]}, +{"learn":[0.3971188557],"iteration":750,"passed_time":159.7471819,"remaining_time":159.3217567,"test":[0.5223473894]}, +{"learn":[0.3969426415],"iteration":751,"passed_time":159.9568284,"remaining_time":159.1059942,"test":[0.5223455935]}, +{"learn":[0.3967767541],"iteration":752,"passed_time":160.1676178,"remaining_time":158.8913818,"test":[0.522416031]}, +{"learn":[0.3966430616],"iteration":753,"passed_time":160.3812889,"remaining_time":158.6796306,"test":[0.5224128383]}, +{"learn":[0.3965085503],"iteration":754,"passed_time":160.5898876,"remaining_time":158.4628693,"test":[0.522426846]}, +{"learn":[0.3964347052],"iteration":755,"passed_time":160.8030799,"remaining_time":158.2506501,"test":[0.5224367831]}, +{"learn":[0.3963336303],"iteration":756,"passed_time":161.0134762,"remaining_time":158.035684,"test":[0.5224470793]}, +{"learn":[0.3962247377],"iteration":757,"passed_time":161.276569,"remaining_time":157.8723143,"test":[0.5224621246]}, +{"learn":[0.3961006589],"iteration":758,"passed_time":161.5525896,"remaining_time":157.7213029,"test":[0.5224556994]}, +{"learn":[0.3959193209],"iteration":759,"passed_time":161.7744501,"remaining_time":157.5172277,"test":[0.5223899711]}, +{"learn":[0.3958587869],"iteration":760,"passed_time":161.9763092,"remaining_time":157.2936826,"test":[0.5224011852]}, +{"learn":[0.395806678],"iteration":761,"passed_time":162.1768387,"remaining_time":157.0689068,"test":[0.5224501123]}, +{"learn":[0.3956605196],"iteration":762,"passed_time":162.3888916,"remaining_time":156.8553252,"test":[0.5224236534]}, +{"learn":[0.3956133231],"iteration":763,"passed_time":162.591066,"remaining_time":156.6322312,"test":[0.5224197823]}, +{"learn":[0.3955843502],"iteration":764,"passed_time":162.7892177,"remaining_time":156.4053268,"test":[0.5223829074]}, +{"learn":[0.3954913307],"iteration":765,"passed_time":162.9980891,"remaining_time":156.1887694,"test":[0.5224311561]}, +{"learn":[0.3954052044],"iteration":766,"passed_time":163.2092466,"remaining_time":155.9744169,"test":[0.5224181062]}, +{"learn":[0.3952516509],"iteration":767,"passed_time":163.4202254,"remaining_time":155.7599024,"test":[0.5223989504]}, +{"learn":[0.3951503647],"iteration":768,"passed_time":163.6224045,"remaining_time":155.5370321,"test":[0.5223783978]}, +{"learn":[0.3950491577],"iteration":769,"passed_time":163.8250704,"remaining_time":155.3146771,"test":[0.5223923656]}, +{"learn":[0.394920589],"iteration":770,"passed_time":164.0473307,"remaining_time":155.1109003,"test":[0.5223688598]}, +{"learn":[0.3948132547],"iteration":771,"passed_time":164.2597507,"remaining_time":154.897796,"test":[0.5223993495]}, +{"learn":[0.3946939034],"iteration":772,"passed_time":164.4804313,"remaining_time":154.6924625,"test":[0.5224052957]}, +{"learn":[0.3946075394],"iteration":773,"passed_time":164.6930455,"remaining_time":154.4795233,"test":[0.5224538238]}, +{"learn":[0.3945199869],"iteration":774,"passed_time":164.8977932,"remaining_time":154.2592259,"test":[0.5224614462]}, +{"learn":[0.3944517409],"iteration":775,"passed_time":165.101782,"remaining_time":154.0382605,"test":[0.5224399757]}, +{"learn":[0.3943477608],"iteration":776,"passed_time":165.3200429,"remaining_time":153.8306191,"test":[0.52247713]}, +{"learn":[0.39426314],"iteration":777,"passed_time":165.563564,"remaining_time":153.6463923,"test":[0.522503988]}, +{"learn":[0.3941691696],"iteration":778,"passed_time":165.7815574,"remaining_time":153.4383863,"test":[0.5224891023]}, +{"learn":[0.3941472749],"iteration":779,"passed_time":165.9761419,"remaining_time":153.2087464,"test":[0.5224936518]}, +{"learn":[0.3940587452],"iteration":780,"passed_time":166.1806252,"remaining_time":152.9883092,"test":[0.5225751438]}, +{"learn":[0.3939841078],"iteration":781,"passed_time":166.3847448,"remaining_time":152.767579,"test":[0.5225697562]}, +{"learn":[0.3938981928],"iteration":782,"passed_time":166.5875195,"remaining_time":152.5456596,"test":[0.5226010041]}, +{"learn":[0.3937222427],"iteration":783,"passed_time":166.8031116,"remaining_time":152.3354947,"test":[0.5226017624]}, +{"learn":[0.3936012539],"iteration":784,"passed_time":167.0191591,"remaining_time":152.1257309,"test":[0.5225416611]}, +{"learn":[0.3935053292],"iteration":785,"passed_time":167.2287854,"remaining_time":151.9101181,"test":[0.5225556288]}, +{"learn":[0.3934463798],"iteration":786,"passed_time":167.4411206,"remaining_time":151.6969746,"test":[0.5225448138]}, +{"learn":[0.393281628],"iteration":787,"passed_time":167.6560095,"remaining_time":151.4861406,"test":[0.522550361]} ]} \ No newline at end of file diff --git a/main/train/catboost_info/learn/events.out.tfevents b/main/train/catboost_info/learn/events.out.tfevents index ff1c270606e67bf555d065ca4791fdf652684b0b..4c7fd6c0bca60a96f09898695e612860c68a100e 100644 GIT binary patch literal 43210 zcmZ|YcU+F|`#{=_@>CVs6lB#;k zJBnG^wbQlus+m-s^X?$3Eel<3_^P>7<$r1~D*c>qc&8|Z8MctBz_vvF2vE8}7E>*y zD$k;wm{riZ!5qG7B~|7@xA&hQ2$E2vok+DO%fF;oSP{>@iy zr7HKhwwM(Wvg;*ZX-idET3b;K&vSdqS2|LaJdCK7D|>w9D_yAya%m%Gt+o8-2TCqX zqn%VWEov>Q*6TjT@>P4OIS zzVxJOiAzf{%l~Ng1-|MeRZVLXr55inl&?BV)tVhG#H?NGy;k#;zEo*sG#Ax|DOYau zRTrsRYetlF*!*{(l)^N+O4Wf6&BUyu?V~FAs+&~Rolex}fG|zI>Mm6aUu%h3?@QaL ztN=oVJ)}x)7E$eItzQCJvR^g#a!;vh`KhUxwPEFVAHM1(RpVC>Wo~g`CtvlJs-hxI zF>6TF;0V6zBURt6iP|`5-EO`zkgA-9M5XthuoRSBm_}cz3U+NGR%j->^pLL%rK+e4 zQGEuyZ^T#qq$()1v6wX=!Ci;1jHJq722uOk>e=#Df2q>VZX{;4-0OZHltP%s0IAA7 zMAYH|EC1oEfl@U`pQwMG^b+}MkW?*=ZYWkb=c>mhz8Wl53)d2*(dyPPP)eb%AySpt zkSMtSc~;>tsY)oXFJ{d$+o%p%vfnlL z@;_3w<`Pl4PV@09Gc{bQBIgq28RvA9pEW|NTBs8>^o`D4z8Wc2UvAbD`zkUDSJ`6= zeaWRNdl^wV2I<<6B^MMtwbgD(RO0?s)_gTunw9Xfu2^AJ$d`S5HAbrT+7M+ivxNp< zjg_kU!-&c{?ba8RLg>pxs+#=L5GxFG9{PZ<#!1zO%S3&gJ<)@&#!J=gl|*Tn9e3cX z2~y?NnW$6IVGH?cqEwAaRu}tv6=9SHN-0cZl2lFfCTdCk^<#WBS*o<>5vBgPuNPlU zk*ef2M0IRycZaX0N>#&;b;Q0Z&rNE~SJR}bi#t&TKkbfyk`>mh0Mn%^b|F#oMz5&k zs~J*N-JYno3n%>HtC>=z_+DG=YgYci>wGm!sd>$vyDx51-<3Fg0)9^j%Xvoi+CslJV5S1J5GXRuA zn8tjm8oQn-J>5D9e6>KTz8Djgt@n2-U;QgpPZ|(4$WeVTUoDiXv!82;(WXeuNR#YMiM8l&q*`1z0LoryPiy7qkSQVVPPcRjU^fm84ar#?M+V zRYgWbsn1z@kgrxq)#Ca@g__&c=Bt%bmG-!L#Q&YarYtVNwO1}oW0h1H>>%o^VL9$# znOZGX_q!408@g^BRLImCsp|5vO02L^!iW)kwN|R;Iuq4mP!jH=6hdF?q^kb}qVkPB z@MMXp^-{IEvQn%tWODs0P$5$rq-tgeQ4eK@@JWiPjZ*bw5m9<~zJBIsZIUYEHbfbg zX_kUg3e(suRr0tBv9F&~+gtF}7OB!dMpS9^o-KT}RjL9f5@oigu8gm?NtLGtQNC@L zbml8dsj7JTSL|z~T5EiEkQLXg0NbVN%weLmKP^2BSxoJas=FpcMb0dp&sRI8>Oy^@ zYT4by9W2Y*B~`jN%f-G9==$I(7E`;Ws@^W5K33nrQ!KeKjXhFzXb@4oHT6~Y8-m&^ zRc9*8#0q~sIpPS@U|IX5svwG}mAA$`;j8^p<>pM(LxaZnM6VDkJRnt?=0ruUZJrHT zOdXV}Zas;bzM)NEH})p${tmDGGWEr?{4BVTVNNo2>a1YV*sBW0=D;ai8vA$%N^EyW(RDy zFJX5z46OL(6+N;qMiW~t%jRtUYF0QSw9u(GO%ZBQg*F2H;i5*B~7 ztOK990*e_y*eF>dew1078?b-W2uu8&Fqog_4(#vOB5{&xK_MN1DTR@o1g3M3u(Ac# z_4v#KSmp`BES|?a<+D@3maiksdEVF*K06I8VH9Cj1>a`y*%@H(ni1A0&)6TBtfXd9 zI19`*uTY$1c3JWqKJx_TbeFK>&P|8$nHR9?qlDd=_ia0$c>{A%Od&A2iW&f zgmswUkPJ*NjN}}!)!Kxaj}AV`XTHGZ7XB6|8FbWc8lU+A>k>=Yii!uRe0CmKS`cAN z%?gJDQwY8J14}$W*qQ1nI(&8kSh59S%JZ%l`RpRF?sCGiH=j%7vrE9DI}(5E*GBE$!gbiC7j+dLI1pyoGLD=fB7MJ;H z!N3An6Sn(FWGgj=AGG8exV8M_6{ zYBFI1>Mgy_PYVU+)Q7N9!(BS@*==BcO$eJ|q;UwCLg?)dFypj*@dD&qSOoLgU0|c5 z2phZbA}+5iEeu#aZ^DKT-WbVGy9aEG6=D84|Kb-FOS=zj!9v0Y$!mXvG^H?-aA3WL z5oX$KuM3|&0A`>~n0t$A{1Rqq5x~L=^27_!y6xpg{Ip15SK|ruJ@IJ^pDBTv-XyI0 z;Mk+UWMwsr!b4!iX9+Xh-s=pXJp$HX7h#Q?4^7~+C}8XUC2UWXA%0P@A|C^5H;%AB zOLrra3%x}HTit`OTBc9%(TK4pz`AM@Hn4HVM(B;Pr@-R#bH&U3*mz4_K6?hN>3hO{ z>&wIV>^ZQLHwYWL^?F-i3SlHMz!ta@_F}NvSw4FK?8Z*QrbgA8$!9NtIWHot??EpO zK8ppGH;S-E&4d5)*(+d2x)RnYkca7$= zcfgKV5Vo|u)_OjB5A5zJ!m1~{#wRLP?<%&OTyHZS)chV8Ca?rVVRZZ4EZbt*ql*>>A9T63&48&25d$@!Y1lO zt9<1PBS{4o)`+k_tLF58B9#J51Geo?w)iQ@-MH&1pM3|`Hl45oS#!<6&mILaQK zs0~c^x909n2WE1cFuTj{n8sKJusJ@2ISzJh#!t%xrr|(X=8+Mv_$&+9)7^xreRfmX zUdGWHXgxe;NjPu5p?$`{gdfd#fAtfI8B z$`T~7JYfEH2%Fo)b{@aTd|=a3v&7G&M$?9P2PlNJ0$}f766SON06xDl_6yjzYlN+= zvlvf98T$>a)Pt~?KgABv8)JpQ8g3!1ZguJ}U`nC4B4BS85%$h$MngU;2DW+wOYFFObu>7H00 zm|Pf16|jp73Da+6f}38(s)1>ZCv3ZgO#wem0e=)0(wneh>nF+i%nDegHeorpvKs?a z2)$VY)2Yf3KPBhXEz<3IKjN~vd-#vt_ZFUv+C5#;b_F^Srw>D|N<)E~du3_% zz^?ZvY^dTmzBOXZ0hmKO!dh=Uv;&HiRn{yD$AFzJPZvKWUf~A~`0O~a7e5JmKmN^s zFh^kRUlFEXHn5tX<^;?%f-udk(e}XPLT}E%e)$k)hhI}L352f}VIFk202N?{~tfOU-~ zEX8mS9!E2F7TC$#gq6E}xWiBL1opy@|h2?()olvHB|X1v$S)-c8?~^X@6V1ZjAW?TWm%Ve)tJqkJdhSsb5T1UCK!VeQjQ zJo)Srutzru^Xv0!J}`wak^o?}&J#AyzGDWT1p-@dO_(%s3Bny!@Abtv#Y?u ztJB0!$;_|4@UM8yrs_wKDz^~ z@DO1}IELO}XZL_LouvLkK zxw|RxQ^J@MnEW+iW3Tj|!cTh$EF_As!e>dX`Roy}>U)G8U-@PbpG5)Fy-wIx*GPQE zk;yLq@4p7}7+7ooVF4c={eU#aqJgFO5N2~Q9XC&mJpm^3AS~WWC!L@66xd{Y!U}%c z;vK-!o&i(eM%cfZI?j+L7e?|NSogJr)z%!Oa;6}#7+^Z)ghkzT#owy2v=_kgCJ|OU z?VQSqp^)|x*wP_{z1?nX4ZSG@77MJ@fH3)($G!RN6|mcF2d(3eD)ET^LE1i2^;W*&k}*n+ep|I=P}RuED4z5T*BtWoyLbuxzO7uU;}y+_Iq5U z9;7k$8JMp=VM&F<=JDAVV9nYSX8S2MfX}`HTiuGV%or&9eiKHL z0_<{Gig=p_rjPp%_6^wPUxdZQuD}IPA*7`OGs+;Wc=cUv=#8;7V3WTP<~!E{uN!0E zf&F?{mBnCep)UtTSLNnn7=#)OfK}62h6<(VSQ7MRPtFqFzt?n)n6L^jL!;yU8qbJ z@3)s*b=L|Ay7 zIX;`Rv@&2VR}(g`;7<-etsGc_Ibq9g`Qw8aOZy8fa5iCo+(Yn(S+d$Si$Vpk=TiwA zai}}?##kjVy`h91e%%KTzZt6n*1Q8@gAWDZ_Zwr?!1}c&OzXs{_Ap7g(3_&R^v~wZ zwFrB=KUL*aO<-2Qe#r=P*}D^ukXV{EFwM-b;@v&*`tp|iB5i<8{76`@#+lxHW(zDk zj<8!@ZsQk~Lg?)fumjvn8yJ`~B=VPFYo3G?4!h1ZR-Bft!f z6L#mismh6=(A!a98;%gxV@0IO&Rk&jzs z+y#W$733uI*>PZhObDxE{sp(8EX@(vn;wK69%q2Bl4W&j76m6@S341A+^h%g@ECIj zcC-#*9tD$Beq$)4od7mdM%dQ=ekuoz0&@YTo&80;-#q>){RWei3(OUm!*{|KwW(;z zXKujU6A3dO-lv$)+<|!|5N2*1f0@rt0@IBn?4HLC+&r<~Jb(u{XHFik)FWj z+Y;tDdl4SlDupyJV5bidc4U`hAfz$o4eXXBVWsO|T;MYwU=DK#tM2n14^dg#Ibc1< z5GF78Rax+a-h6>|`-d>k(|3A6kuvp~MZpi4^L4}R^(-1+wK$g)%*DsNK*(S2?FMJo3L_?)2sL_7?}P= z!n}6=vzyPZ05f$bEZ1@IXg&)8cHD}v9|P9lIT!2gDzNOWgq8L%oCs-3VI84%`E(quayTHQ-2}E|B4H(N*AiG@9i!yXKljH z`+AQ7CKpB$2JA~K!kise;v+3%_kc}pLfFI&eW&r$?gM*P_(@!XE?FnzVH`^f2lnbC zVUY(HSRzTV!i#%D@ky2l8+ z9k2+$-&op1U_M6(n>A4FKhhonTYHGG0ljQAp-81Lk|@l!EO9|^* z;NZniiw5RrM%bbyN__vu(w+coJfE<=kN=M1r#%H$J)W?ek;|=s$?DduCC`A>ZBN*T z5zhD|%-C~aI;{y)>uc76pB4km^ly@Q;~ak*i(iW@?FF#0>4aH-oQG?PTv@f^DF+`;0HzR%d<)FZo3QiNHc@;Q53KSuVQDW<;eL{(B><~uMcBT5 zUGUGZ7<&io<5t4r(l5>67x^BT;|9XaY?tDBiBcHJ2VkG)6SiWb<_Ab)>?5#2;|bg7 z+tZBC5`hgJOxW{&b%*j<60omEgl+Eh0gwJzk)MDCYZG?K%vc6#vU)XZ$!B0*T7<0} zH3Pr98T$h4Nj<_Y42{R1ZZP&0*erFzY;-K~9Tj8Az?!NN*1q|fIsD#IfJOdI6jv_) zVF4$B$%T=812(vjunT!@+wfT`vI4>$+B@#!vov6#-w3< zaSz|cXL-PuEF{dJ_g$3}!}>LgLO!tMQG{K(JI)T$WCAMymfMf8x6z7&eD(`ir9NQ; z7LD+?ceG-7lh2DyQeaQYOt|f6#8-3!lVqlgjge_dG zQJc^H0P{~I%yGcpBYai@Eb2L7=bd(62c{5uD+Si$K4ID`9^(?kSQ#+y%Y-$LuJeqa zRt{{~CBl}4rLEz!zreb<5N7ox2siYs$O>SA)`S@ZYu|u0r7)68U`6{0OIguaYnCWGF-O88^Fq4slx%R)%2a06O7Fe@>gpKk4f_r7g4gs5BK$zE_xWoK3 zJ77g!2^&>4@GUU8Fp|T-Tsjf9-t(mipB(`vZ%>%{prLs3#L|ueyW5Dch$W8a_-XdQ z+Le9~SFVRg+TjU?Lg>u_*pzg_hCF|za+?b57%e}lte!B1?H(kSajyUseI-KEVwmc!Q<>=fGLHM zoCh|c9$|Yn+;7Wg{=l~WeJ`%cjfOqPPYFxA0BlJvVf|+fY|c-+2&`QqVGFAA@x`^Q zQO(_b3D}bu!cN;&mO~n20l>oV5%wu>8=mws76@#wKVf&ae8+c-j9ms6=1!R2w2i&_ zMFs(TafC37L+W^hBo{^!46I}?VV8Zr;#Q5ZE5OV*5Ed4b?*~OP76NSHbixdC((o?{ z8M_MX-UPy?Y+XGY(iB2(*MRjLMp)1a-EDk!9oUP(g#9u6)`HJ&02|+rF#QdURDLQY z6nPU^13kiSpXh7BPrC)IPXofzZ)_X_Oev&=0;?!|C*G#-Q~rM8v)jNNe-JieT0lcS zy94aaH^RDg9r7=q-32CpL0IOmpNoLW8rR(2VZct^Cv5O(lhJ&3514^JVJ8yKedM$I zz{Z{??8_o&JZfh}h68hUAgsH0s>+#y(AxuG&khmxRn~qM6e$;21hCkhgsCZ(^yRZi zV7s;ub}zo^Ni_svT_dkieNH(~3h)mQmdgpd{u%(**Z_PdgC&&7&-0xYC4VKVcV*qc&FdkSoa zI$<6?Gx1!Rv1h<$$Otp&zwaoFgt6zqh88D?Ye~t)^SAgc2H1;K!rnajGo8;~0E>7> z*wc%L@Qs+PNzJ105}40j!uC(rzY1xL#RAg|CG5cv)ql@iNP7kB`gy``n-9PTF_snw ztlXV2Q~gVLcQf`H*ck`HVoNg9pf|Zt z5N6VN44%%gA`^g>_9x6$Z}Jp=+B;yrT?va{*|h+eLKw+=VA`Dt>#;d~5TAVjHl{gY zHx0Xo^VvsWxwQ#f^jwz9XNkae{*D*#x0}lhKLb+=y(Iw~ok3Xt*dDL>>=Q8WG{T;? z_=Jx}EbTL}j?V~dt!CPopY{dVrpJW6lLz33i=}-9wmXclf_`)MLYhpoW>H86cFvEm zc-zYz_$&q3TzA3><2$_Lvv0tD+Ypu-)V(L4r2<>Mov_7&8t>$@G+?v05N72Zk4Lt0 zVI<#yZCys#E197oq%rmb*g6ZsR++fr`6OfMz$&H@R@|sh7(Xoo*p*3y<#oQ(8JI%o zEfZLy;e-wJ()Qr9EMPW9giTtr0iWMkS~jqm?Fh@w_)*PI%K`STEn)s`2Xy4KpTMry zCTwunaD1dy3M0t{cD($pxRyjs8Sxj=7|R1TK8vu3CMVqaEFV}x3SpkjuHzvpODh0Y z5=YoKyYtSFCTm)AcmDzw6hYXDf824a#@KIQ=GO@8JF!(EKdlg0z!kzac*$?`SrM?y zrwPmKV5#zqBlK1bEZLE;-MZ7aLz-M*e}H}4OxWHXnt6Oy0?c6nVQ0=n%;2+9V8ufS zD|@{Z593&oWx!Gl2-|e9Eq(_mgx<=5eU%Z`dRpV5P$Xl2fz8c%Bi;dHjOyIsvkGAQ zJ`-j#e-55Jv9wBH`(F~K9$t#aV~kY+3ymafL1t~0ow+cQYG5625cb3VhynDb6qrI? z`uA@#AHw$b{i@ApR=_%)APXJ*D zH?JMQPdfo@gC}8H3S+zgN?{}}!0K8P)^S@ml?}bXT!DSHB&^oF;kcb)X>Pz$R}j`j zd+7-1jWKs%L30UnG=7A?YG>>uu#RI1(^b2Vhu^YhHH(4=u*5!uS=LtT4n;C{3YbL) z!oD_ZeU{Ho0~_);PW()Ev3Q`eatTGA0k*k_Fykq|Rkkw%I}0o}jj($IrzAm5vw8`OFKLSps1vxBB5T4omX}77|U^yOMWh{4^h6=WY--qiQ^!y()y>&H?*z zk+5pZNKZ&(%oo@}55mScZo+p|jQIfzJx18x%;9bLY3G3*JV4k=J%ur!`2(}xN?1b1 zW({CUVI&uT^;kt%m!Nid_R82rV8)9H^9{Ex;HO;z7C4KrY}-VYB}gbT0NB{kgcV+x zJc^$d2yFKt!YbUayapy~Ub84%1{Ttpu-j)3SMga8uuEEmeR1zl#Am_4QmS5wi^3}H z40}Gi0&HF`Vf6=Q_yUs)y@dcf@rJOGgOWA*>?*L`j|rPUKBf+zT>~~Gl&}f5^YJ!i zMP3JXE`+eaFf063J7YJ1ZTBOrqx-(NCJ%d4bs>afkz;s>`rd}rv zzj5TkNFD-Pb&aqF=~q<#tUzFofazT(ENQVZ{#=%&MFAV^O<2?4SMiM)V~>GdK10~* zkc!qYNyegqxtt)Zo8624z!XAnPk>FgA}k_l4eloydkXC09>NAD7ybu(25iA5!Ulgo zw}4;db6}sA6IO510DQB;ii`nv>|et6n)JO7X-Z)vFMwT|Kv;v-6)Hy&0(%Lp)nLNn zDvsfS6-$c+mZ3*jyP!9h_(i?~X4#Ces-yA9WG!nJg*afVs$PnVLg4Xf?;(w`*TD8< z6V@*v2j9Oj_6Atr6vDchtGxbU>@BeNZwa&ZG{7I{G8PZ4^>f0Wojtf7dXozyNdQ)P zo3JjQGq>^CJ78mj35#=E*PGAY1N(7-FvGL@cwWMa`~b|^p0I0Kr`vz=&5n&G-S>vWx)~aSvNCnpU7h%P^i~V3EjHLnFmqFOBBL=wXW$Zh! zqECdCe?B>kpY{XT-FU+OwGBT6OfK}64s5|Q!d|DH!H+Uy8Ne1s6Lz@K)TR8iOknjw z2^*)pgWDOFmIbWAdBT=$*2J?Q#I>uE~Z`qK{__(kRe)B1<7Q|2}!fhmO|3xJt+ zB~1OGi5j2%0v6kXu;9b(Pw?4qV7+P+X8br@WeF0BECjZ{EJj=u23v<7gfv;}n!CFQ z*odEm-Llto=Cfj8rC$ge+x!!ry|T1Fz(QjQ(~CNeU&4%)04oS3?E63iJjG_L6j-eg z!gh_$ErZ_V!br-1^}a-y(WPvB>|?APSRFUQ)K^`<%1`?XEa5m|DM`Ox@L2`0diI2E ze)qx-m_q2S64;0Rgk{=}#S;w1s(>BeN|@2W`gn@XST(Q>iwIl4Xwpl5kqQmz`?sz$ z39HUu{FBeDfTfNhY=BoWe&Z;Gkyrz(*Pk%0DS_W1jWHWwCHjPQ+UJB1VvN}Wo7IxA zm^T|@_-TiL$r=!L@Znqht1Xsh2h6VOxwt48%`(Fmpt3eKi^5@GZpDPz*fdOrA{jdZ ztT2PHgWJYF;IpH^K7S_cj#FMHpV2&v$*#UTd$t*XHWVgntbCDrRW>&%dsG*5r*r zpSF#38mUj;YQAafjvcMtW)A#k^#A*>jdjDD^k8Pk%v{<3H!bQ&Rh^;wqVjW2y2w{` zrRt{M08zdB*dmOt>PeM#`~ISO*zVi{zN#-(9Xs|D)w%9FmV%Pk_M$}tsS0b}S5(Fg zzIySMnp7D!?jx$`OHWSmmAX{9x9BY@*)yLPeAQ5@UU%&!s*wXX9|5Hh#-$-u>VtZU zDn+J#jISC=mEQCoqB=V9mK$F+ma3U6^hDLBNsBIg)kLax+vtkQbE^6xzG^B}Cg;10 z%3{YjYrbkGRl^dxiE47o;?JOzLJylum2r7jQ9bGt-IA|bNL83VQ9}n@*uhttQq^!x z7cpx>|F_@xN=vE&{5y-Puqb~aU$vB~Z*Mz^s_EqX2cTrD{_m>MqLowywI%B5M>h?= zYAsboD>{lD*4_U_xP%fR0U2TYM#>&Jy3F? z8*Qbk$gQ22HE8Xb4Sdy3s($6R6;=3!1Ll0yUaGRj5Ow;ctqEUskg6chHe!~~fsJ_e zvT=2ks^;Z7qT28BeKBMygl=?_D&NIKO*F&Z{OI5^uZ859e;E;d#SzV;+TQ8zM z-$+&GtFBVD(zCUg^~&C1JSe3wu5MD*tR7L@YQBWY2x zwfLE)nDy*k;cvd`EmigA5M`(P^naDR&s@0DgixrMnzg^8&Bcy8S2BIFG z9hAgZBc>$oB8q8D5)BJwUJoi9jCl1z8Wo6VGD?AtQgwd;QLEJhKk(IfsnTvh)VL>sR-oj zQ4_yB&4w(6(2a>wr7?!6!2QD?@zo@$YEh~tR%lf(ET69?OI1cNQD;vbx8Fz|mbsCA!UrZTGmBEJw;<%bxhu!37O_eGmf1*Y^?#0~~%bF%t z?G{qi#1Q8fQ`4m?vkg)E0+$s+g|fA^b9shTHOioE%%eH(qORB0n5moRa;5uK;mMZIy^~7FfYJodZ zJ1o0~^3_JE8e&A$pJCrL`D&9?t^4~|tS~@p$05GjELC%_61Cy(ww0iiLN~TZRl;(j zEVDvd^VL?V8rOy>lMX3(|6*DHO4Y=aKjOHKB+kIS4O82sO4pUBukl)V|6*#pR0SDR z)!_C#7?;evb_LiWRj1X6s#;RlpRX*XO7^5m9M{&1j(B%qSv#f5$APGw2K(LkS-Yev zY7$W)Gu)qok_+S7Emh|g1D{y~ zi;N@e*8xQ`pV98>vBGG1h!%fVKcWp&*C#DVEZ%((>7~n z08B2l#2HxE=M`d0Mi!VV_{;^E_XWa!JWg1~XRg4^ED6gWl6s2I+<;A(L|A2yEH7XR zVQfc$6}BZzBiwTepB)9Z?$ZykB~R?4_wboJu<3UQbL*7;gwH&H<+>BrvHk2XeC7%4 ztSMnD+UtY?Qwl9P2JFdD!V*1{rF?cA*x0&+g=`+=z-L~-sxr&PPQE&|Fr3d$0P7e{ zSV+{R=X~Z3?6U`9wz}ubfyvg_E(#}st=ULe;JN|b`OF7cngL-q{@#kE3k23Pi7>5gYA^Wg z9I%g92+R5!`kT+r1ABXru!HSm+<_^DmRtaqy@Iewn~GcW*+pP||RUw~U0XE*7 zu)iVA*4)KKD!HSU^~J(&B(y3gr(gB7F|*-{s09t?``I%-3NC41!0Xc%Cq@Q z32ea)!tR=@?FA+mS`rKF$0@?Hk~*96*#lsW_7FC%pITQwivzZC31JCqJHFwwhrl{d zCM@&LYn3gwFt$g)*7hcBXV`9QNK**xF)-ccgiZ1G)#bDQfTeyb5`XTV=;$RuFJ28506cP0wbWSeSdcQUZUJqYWvZ!@-pu@qo|EeJEu7_7-pdj{;-=R)!4 zew5qw9G|5E%Sa^brvAV9{KkrW4y+`cFuilJxsWCoTJi!|sRv^g_gVnR&|N6 zQ?_NR`RqL~ZCAn!7RKTIjiqG*^RpzZ>*?fm{In0ia+VS{eMKZrQRb@%*)MKYEH>~<5vA}op)@Y8aE?X37BUM2ZgcD&-VkHFe}B5dW9 zs5`*q!r1bFg`^N>**CnB&prX06iHZM+=EYimJe)x0Aaz}%7J|L8JL<2Vbedh#d|j! zTLG{~dkM?@r|&aJQwS~j0_>?7VN34*Y`|w*NGXi17})vxgq_VPmrUdSL#wxxDu#{)rzzkMg$4H`%xswJFeL!OvjNt&I$yj>f=<=A^O-HMTW<-o^6u-$XLi7<9};$P$l12QWLs-zw>>a} z8-zK3-i_NC#ts9Udy=pX|1IjsPjdiv&Wf-GFa3`2nIo|Hb%c!>s#n5iPQbQJBFtgJ zQG8(~7h2*B?D`PGCRX0WZ75?dz}z|#rkvVH8;WGi71*iYpTw(V^SNm_O&N0o7GFr% zz3gqcC@6%n9Rb!knXn0GoO?o%j2#6w_a0%cM)!01%pF)s0AWl2?f#a}Jb)c=Cd_$y z%3VJ51eUjru%IV^QJeBf>fydfAuHynr>36ILEleu2+U z0Mj2r*rc+^JU;UVcDn;%n?eo@0w(*nc2PJ9?DMxg@tS;|*9I7kLWUiAcig$b3A2$%V1`0egCyu-QFio_uy1*nDTgx?OIf@>Weq z^9R;{H(`nMGDh>$0)YKoMA)DPW~qF32H0Lh!d~B+RSrxc6nPfd@_~eDCO0hLvp`^e zJqc^8V;|3F=Yaj&im>Z8$_{*X9@vCBgk7{a^BHIE&+RRgD?l5_xPa*V?n@b0tvGUbz90W@-nc0Z3yc)Y5zoEvTe1qI~Z7m z1!3#YPtoMFE5Oz+BW&RH{`fHkOA7(^$B?iqKZnlYr-cHWFpw}y*O3SL>?*MG#)PHJ zEm;dpF0>>J*qxeO@hVwA#pW-cT?5wl8)5x>&amRM>%dmMBP`S?RAo^RiVO$lnL=1; zbi_%1+6`bsVhH=X^3XY83L)(#Fx~5fRlC1Y*_jJ00@#}%!sgF9`JJB@3CzNuu+^8t zRF3V0v|GT+>!`-Vw zXvu%TjGPIpNWR|!(inRJOl=2YOVxU->`R2Scwjv?5cb{wzl;2|1YjMO5w=XP>jOS} z3an@H|{>MJ57UIg+rng^kttED6}p0fhDTEygD*mX-`GvJGJ-4~mBK(^7!F z`kO7z?y<>TV}Qwa)Gi9ofGsT~Y>>_JhJ2O^>`X3Uz0!;DXp5yi2j-SQSbHs_kNmV3 zzy^gA7X8)`@7*jd4OrtVgx!9reiPE_YKq-7K1&B?cAT()4fpVOV`;B| zX&okP{MnZHeF5L)sEnEnF7bhh5YmpY8S z1va7|VRzefvf!t^19ncAutD_~;#UzY?LDyOod_EdR<@0wmI=&Uo3Lf?T-E?n3N85n zEUyk>xikL^Uew4JbzC1Wf2XKh))v$U_k0!#^; zG}KzS4^Xpn=BL>LGtMAv`j>_p zd}ar%Fom$+uBEt^D20~T13P$!u)B3;?u0bP4g)JcLzs?v4|P6s0G4)&FrQ_!&hwcg zFnu?|OpHVD>oM7`+C{+$*k(tjCljAA5U0#d=%bsltPgwfqBIe zR({V$^#dmlrFRK#`1{26plwVXMsRwB$2?U=fyteP5>3=Cc4`uJZ}=%jt&?nXJe&z)@&05(%bm}}fHyh;>8OD+OyUiD5~OPdvv6SF77*s&Bf*Q$ZU9?lOjtr`qf}sWp(QtgwVg=V+2J|3du1#F znEOA3+1U|KjTpNR%;6GYo=yGnv5zq&unw+-jUIp3ieF?b zumDHG{Ev2>!DkPEt+yxaV}(HoFxlSPwImK$&LP4Mjx)mdT#P*g_G1-c9&IoF;HNzT zc6KRY#zwwXeD)Yv(?x{2k2z`tOfHP=KVUK=!X|v`TZhk{08{ToSo@%@z4#w(64rz?N09NBe*sJJVl@}jES{ks1)`XR;e{9H4dkHLKJ7L{r zMz{--S=BBI>A>7K6J|8<_HRgI>=m%LiwKK9ei(N_jAZ~DHifXc4eY-1(_RA`GlH;w ztw!NAWod7KWez56o31|YC*?v*-U174N7zQoGlftjWAA{S)+Fqx>xO20_8!=yhJ=;W zzvRhhnZV{WAZ$bU@)dmc0a)|8goS;oi+59n(2^`*_o`ltE7uY;ExauxExGB3hzS@^4bVI>s#8JM#tVYf5SEP^!IzS>2h09gJ}!mh{P zUCL))fSFhmR`Y7j6+Zh4Y{)^vZr&>F$!CSY2Ja>;cG!u{d{zXk**e0KH+W10CKp;# z49v`wu+3UaaTmncH(*%?guVIGTV-D&q?G_G8bsKsHcO)UMV11q)+Nj=NKIvfA*6i= zX4-|Yir6*nAWb2#GGMdS2&*xfai7o1fxZ5cA+9C5_As8X1LLdk97id%B;v z-NPw(>tJa$z|QO>>_SI1m9HX%u_@qxRCaM8Vc%Qt*MTDC0<#A8VK!lAbL*S%nGLYZ zhJ-!b@BNa`Y=KSwhp@rbsdf3x4w&X3!d4|k{vI#zE$So#*n2s0+V+l%*y&rFh9)+SX2|jd`@gp=QC$uoqoO&SFYcQ zm+(E8QfP?_uxXzN`&{IqGP?!l3hej`!bY?-ONSy^nj5hBiG+FVX^C%o89M^3;yz*b z{&ac_X|e;gi^5T0Kf(#?KCuz*l^Jsfwl$P6wHN2y_-P)%iY`;OKGlWKJb@kXCM?_R zs5PG*16JfhSZc)xe9t8pT5=qilMP{a(sWfml@gd2u>rt0W%m!m|L_DZn4?eP6LbWPuS0<$x|UsDHQ1s%(o|DgI^io zhgpmT02`=FnEsU1M*Orhz(P9{mj9#V8=sv8=GUIEsBh~&^I0IU5FNs%iU|Ag*f^ZeE&}WLnlOzY@e6^; zg(5Ej`!9j8p+7s|LndQE!0ts8Hnsh8`~eud4D8T(!e-{=;h8XF!N8)t2n)$Ng2z9M zT>UIW%=3}IjEhvK7=QfSF_U{6O7R&%Nn(-;c}W;}>6{jLL(U~G)t z0H)EEu-x~iPJDI~SieSuZ8ScjvT_Myivae!?4>wO@5lVw1!=NFwTnU|F!wyd63=>J zCmFj1EH#_3%O~9C@YAAz$)6Io&SueaKD!ORkAR*kpEyD4Mwz()T_6W0>;67?PYv;<%YIfP9r_K)SW zr@-#MB&@N^eqTOI1m+h-SnJBA{{oW>ElC2_G=i|7vA^&Q24l&9G8fqVZGJS&tXJx>GLJ3QM zQmV2j2x;ZOf_(^^JL$p}e%cRUSw{%dUYWL&&nkdr{!7^D$&K;NlR_x65}3v_8Zs-8DZHU-Ei~7SQW4ZUsJ^ykQ!{# z5Q}Sq-pG;e=iKGXr-) zGTYkOt!N0^O*y%Cnkg$&7ORm z&m4gb8ADi3pLBd3tq{iM1k6N_FgGjjevrnPGq9!Y39~iZ`M)q1UKvfS-08 zSk79)LVS*HBMW->?nR6he`wfFf0+5A5nA!pw#y;r)$`EdbcraKfy+`HX@zrBLJ< zV4p4$wrOZzERwOazz%s6)+F@uB7RyRu$!)g4g5TA9iN>8cIY5smkp!81C!a;&hGQT z9JdqZ=3_XB&n^J#ZBE&BBm4na+C^Za%m~wXko}3Db_rPU9Ktety}`{BOA7)v*^n^X z^@nvJO)j+LGO#_v30pk*ayg#`1H0OtvOBZ!`HiJr0d`T7u=Eq1@trwiA;7Y#lf^|L zXw;xoC{iJeEfm;-BEm{@ug~VQtH9h-2|KbxOXVqFNDBjI^^maA3%0m&u_CVl8xcX6 zX@QG^U*vUQXRZ>qf7gZxU`nCLa9}$F3G-WBg5O{;b_3YTWC^LfTDWTK0tX zTIr8(Fj!gyu;T569aC$H51EWb0_(MkuxINX4?#<0hieywTfjCM6E@a!$9g`C0`_P; zVIOS!XY<)@V1>g7>*KLx2%kj*tLj3S&P+4BQOSj|#QO95t4ohU8}t=_D_&(m4jGhlna5Ef}_F_oW| z3QRwXus`a9viR&durALDTW@m+KiyCWEqMVB{(;7%v$5>O`b20WF*t!0M z9X>jB2|q0p*zJyl1^fu_#%CXZ)z>5}`Pi<%e3k|5Y0Xn{nvPg`_8Ks`(2{Im3(5%_ zWm0&C&vJn2mJ+sZ+As$`%LTSSpRgY3!~62tM_^@d33GC(!P7p4Ft$8kw_XsYxqfgP zNMr01uq96j3o5IxvM2~?`M^vc5jJ7DT@*jFot?Nzr>>Du6v4k0Y@K*9!39xxR2rG4o>BDEGz>-@LX58+nAD?{( z)~_C6{a;P?045h&QU>h6uLN;X`0r1UjL*t}jV~muQJT&aKKlWz;yqz|8m8j?jTKn| zEFy`pLq1-5kfspERtfA|1Yz5}#z*tnPhhgEglT5(#3hKO{Q}ni5@A|a1q=CUzk$_1 zPFQSP3+yCIs{)qpM%dEnH|s%~QfSE^VCv3-L!&d{zz2Z98GE zJ-hDYvl?IGUwXatx%JmzxA9=Se;qT?fJ|aSk45(lKnTUoCgSLHo(+| z5Y{CiWDh^h7FdNYVP3PR?&C8%U~^j&_P**Io*K!8BJF`yYY^sGwG3Z;Fm@Q2V|Bc^ zC<#&J(RFT zUmN3-Fk?r7Rd^9rntJgwjEymOVC@eRwkYTDJwEdQ=CqeE`JG~%rYy}9SmHLqewL-; W$%@RSc2PJ6Y}iu5eus9M|9=3J6n`-Q diff --git a/main/train/catboost_info/learn_error.tsv b/main/train/catboost_info/learn_error.tsv index d094c51..c5fbfd7 100644 --- a/main/train/catboost_info/learn_error.tsv +++ b/main/train/catboost_info/learn_error.tsv @@ -1,789 +1,789 @@ iter Logloss -0 0.6887138297 -1 0.6843872854 -2 0.6802627852 -3 0.6762009319 -4 0.6723192545 -5 0.6682080654 -6 0.6642342664 -7 0.66040895 -8 0.6567665034 -9 0.6530918354 -10 0.6495993504 -11 0.6459855333 -12 0.6426157539 -13 0.6392258492 -14 0.6358979576 +0 0.6887138826 +1 0.6843872326 +2 0.680262838 +3 0.6762009847 +4 0.6723192017 +5 0.6682080126 +6 0.6642343721 +7 0.6604090029 +8 0.6567664506 +9 0.6530918882 +10 0.6495994032 +11 0.645985639 +12 0.6426158067 +13 0.6392257964 +14 0.6358980104 15 0.6326337693 16 0.6296338495 17 0.6265263841 -18 0.6235141039 +18 0.6235141568 19 0.6205433947 -20 0.6175789713 -21 0.6147449654 -22 0.6119898755 +20 0.6175790241 +21 0.6147450182 +22 0.6119898227 23 0.6092233761 -24 0.6065007718 +24 0.6065009302 25 0.6037280922 26 0.601134849 -27 0.5985698655 -28 0.5959878205 -29 0.5935677806 +27 0.5985698127 +28 0.595987979 +29 0.5935678334 30 0.5912109686 -31 0.5888627138 -32 0.5865621045 +31 0.588862661 +32 0.5865620516 33 0.5842648229 -34 0.5820778865 +34 0.5820779921 35 0.5799317814 36 0.5777438942 37 0.5756289014 38 0.5736006422 39 0.5715183462 -40 0.5695853779 -41 0.5676576391 +40 0.5695854836 +41 0.5676575862 42 0.5658429393 43 0.5640453538 -44 0.562331544 -45 0.5605908479 -46 0.5587755142 -47 0.5569407421 +44 0.5623314384 +45 0.5605909535 +46 0.5587756199 +47 0.5569407949 48 0.5552666016 49 0.5535532672 -50 0.5520047375 -51 0.5504471752 +50 0.5520046847 +51 0.550447228 52 0.5489185594 -53 0.5474294544 -54 0.5459806526 -55 0.5445916453 +53 0.5474294015 +54 0.5459805469 +55 0.5445915924 56 0.5431495519 -57 0.5417205055 -58 0.5403851653 -59 0.5389834808 -60 0.5377107875 +57 0.5417205583 +58 0.5403852182 +59 0.5389835864 +60 0.5377107347 61 0.5364798765 -62 0.5352338583 -63 0.5339945486 -64 0.5327745188 -65 0.5315706526 +62 0.5352337527 +63 0.5339944957 +64 0.5327745717 +65 0.5315705998 66 0.5304691555 -67 0.5293499101 +67 0.529349963 68 0.528244557 -69 0.5271158037 -70 0.5260443622 -71 0.5249990677 +69 0.527115698 +70 0.5260443094 +71 0.5249989092 72 0.5239809237 73 0.5229378476 -74 0.5219464315 +74 0.5219464843 75 0.5208603056 -76 0.5199106717 -77 0.5189214741 +76 0.5199107245 +77 0.5189215269 78 0.5179007418 79 0.5170257982 -80 0.5160979269 -81 0.5152054464 +80 0.5160979797 +81 0.5152053936 82 0.5143571779 83 0.5135479977 84 0.5127352784 -85 0.511893507 -86 0.5110974267 +85 0.5118935599 +86 0.5110973739 87 0.5103353109 -88 0.5094643818 +88 0.5094644874 89 0.5086953991 90 0.5079769143 -91 0.5072723744 +91 0.5072724272 92 0.5064938838 -93 0.5056995994 +93 0.505699705 94 0.5049644228 95 0.5042152484 96 0.5035045284 97 0.5028365414 -98 0.5021301528 -99 0.5014404559 +98 0.5021300471 +99 0.5014405087 100 0.500777487 -101 0.5001429892 -102 0.4995491114 +101 0.500143042 +102 0.4995490058 103 0.4989485253 104 0.4983499992 -105 0.4977427575 +105 0.4977428103 106 0.4971949405 107 0.4966445881 -108 0.4960738992 -109 0.4954706192 -110 0.4948638001 -111 0.4943255439 -112 0.4937688 +108 0.4960739521 +109 0.4954705664 +110 0.4948636944 +111 0.4943254382 +112 0.4937686944 113 0.4932566907 -114 0.4927494411 -115 0.4922379656 -116 0.4917602435 -117 0.4912688932 +114 0.4927494939 +115 0.4922379128 +116 0.4917601378 +117 0.4912689461 118 0.4907662919 -119 0.4903079026 -120 0.4896477861 -121 0.489197056 -122 0.4886972542 -123 0.4882628989 -124 0.4878229445 +119 0.4903079554 +120 0.4896477333 +121 0.4891970031 +122 0.4886972014 +123 0.4882629517 +124 0.4878229973 125 0.4874311637 126 0.4870381152 -127 0.4866496622 +127 0.486649715 128 0.4862709813 -129 0.4858681607 -130 0.4853899103 +129 0.4858681079 +130 0.4853898575 131 0.4849540232 -132 0.4845096845 -133 0.4841039059 +132 0.4845096317 +133 0.4841040116 134 0.4836538097 -135 0.4832327127 -136 0.4828748437 -137 0.4824942083 -138 0.4820718964 +135 0.4832326599 +136 0.4828747908 +137 0.4824942612 +138 0.4820717908 139 0.4817320925 -140 0.4813566338 -141 0.48099718 +140 0.4813567394 +141 0.4809972329 142 0.4806665143 -143 0.480273413 +143 0.4802733602 144 0.4799522024 -145 0.4796210085 -146 0.4792147017 -147 0.4789183174 +145 0.4796209557 +146 0.4792146489 +147 0.4789183703 148 0.478532664 149 0.4782636944 -150 0.4779824172 +150 0.4779823644 151 0.4776756799 152 0.4773818311 153 0.4770647406 -154 0.4767062377 -155 0.4764296089 -156 0.4760174917 +154 0.4767061849 +155 0.4764295561 +156 0.4760174389 157 0.4756622109 158 0.4753690488 159 0.4751120698 160 0.4748108259 -161 0.4745126986 -162 0.4742214381 -163 0.4739676812 -164 0.4736240741 -165 0.4733877484 -166 0.4731457709 +161 0.4745127514 +162 0.4742213852 +163 0.4739676284 +164 0.4736240212 +165 0.4733876956 +166 0.473145718 167 0.4729062759 -168 0.4726252629 -169 0.4723725624 +168 0.4726251572 +169 0.4723726153 170 0.4720965675 -171 0.4718570197 +171 0.4718569669 172 0.4715016333 173 0.4712788301 174 0.4709921123 175 0.4707643438 176 0.4704056825 177 0.4701234017 -178 0.4698584465 -179 0.4695728908 +178 0.4698584994 +179 0.469572838 180 0.4693260536 -181 0.4690976513 +181 0.4690975985 182 0.4688616426 -183 0.4685305015 +183 0.4685304487 184 0.4682865695 185 0.4680025456 186 0.4677844436 -187 0.4674939226 -188 0.4672482474 -189 0.4669966563 -190 0.4666803581 -191 0.4664588227 -192 0.4662119855 +187 0.4674938169 +188 0.4672483003 +189 0.4669966034 +190 0.466680411 +191 0.4664587698 +192 0.4662118798 193 0.4660336056 194 0.465759565 195 0.4655335925 -196 0.4653787184 -197 0.4651595599 -198 0.4649592589 -199 0.4647202922 -200 0.4645238472 -201 0.4642532401 +196 0.4653786656 +197 0.4651596656 +198 0.4649593117 +199 0.464720345 +200 0.4645237944 +201 0.4642531344 202 0.4640300671 203 0.4638038305 -204 0.4636120338 +204 0.4636120866 205 0.4633936149 -206 0.4631845454 -207 0.4630175222 -208 0.4627118941 -209 0.462500342 -210 0.4623026821 -211 0.4620509325 +206 0.4631845982 +207 0.4630174694 +208 0.462711947 +209 0.4625003948 +210 0.4623027349 +211 0.4620509853 212 0.4618490996 -213 0.4616663883 +213 0.4616663355 214 0.4614388839 215 0.4611913601 -216 0.4609538723 -217 0.4608258318 -218 0.4606162869 +216 0.4609538195 +217 0.4608259375 +218 0.4606163398 219 0.4604787385 220 0.4602659714 -221 0.4600531516 +221 0.4600530988 222 0.4598740322 223 0.4596930112 -224 0.4594828853 +224 0.4594828325 225 0.4593606024 226 0.4590814381 -227 0.4589115626 -228 0.458786797 +227 0.4589115097 +228 0.4587867442 229 0.4586156537 230 0.4584218498 231 0.4582468505 232 0.4580919236 -233 0.4579602912 -234 0.4577697094 -235 0.4575785994 +233 0.457960344 +234 0.4577696566 +235 0.4575786522 236 0.4574154851 -237 0.4572040914 +237 0.4572041442 238 0.4570666486 239 0.4568047043 -240 0.4566408504 +240 0.4566409033 241 0.456427608 -242 0.4562375017 +242 0.4562375545 243 0.4559405892 -244 0.4558015617 +244 0.4558015089 245 0.4556457896 246 0.4554681492 247 0.4552844343 248 0.4549960262 -249 0.4547091499 +249 0.4547092027 250 0.4544524878 -251 0.4542734741 +251 0.4542734212 252 0.4541751195 253 0.4540318662 -254 0.4538242758 +254 0.453824223 255 0.4536671832 -256 0.4534657729 +256 0.4534658785 257 0.4533309183 -258 0.4531104921 +258 0.4531105449 259 0.4528481781 260 0.4526748163 261 0.4525394863 262 0.4523340087 -263 0.452107508 -264 0.4519560673 +263 0.4521074552 +264 0.4519561201 265 0.4517016766 266 0.451566241 -267 0.4513973162 +267 0.4513972634 268 0.4511307236 -269 0.4510252909 -270 0.4507913422 +269 0.4510253437 +270 0.4507912894 271 0.4506708024 -272 0.4504389667 -273 0.4502753241 -274 0.4500877004 -275 0.44989284 +272 0.4504390195 +273 0.4502752713 +274 0.4500877532 +275 0.4498927872 276 0.4497752582 -277 0.4497071707 +277 0.4497071179 278 0.4496092387 279 0.4494378841 -280 0.4490959673 +280 0.4490960201 281 0.4489082908 282 0.4487697387 283 0.448658654 284 0.4484694984 -285 0.4483210158 -286 0.4481748573 +285 0.4483209629 +286 0.4481748045 287 0.4479443949 288 0.4477727762 -289 0.4476346467 -290 0.4475678269 -291 0.4474215628 -292 0.4472752458 +289 0.4476346995 +290 0.4475678797 +291 0.4474216156 +292 0.4472752986 293 0.4470165765 -294 0.4468634455 +294 0.4468633927 295 0.4466839564 -296 0.4465560743 -297 0.4463076524 +296 0.4465561271 +297 0.4463075468 298 0.4461113659 299 0.4460250547 -300 0.4458235388 -301 0.4456807081 +300 0.4458235916 +301 0.4456807609 302 0.4454037096 303 0.4453104787 304 0.4451840757 -305 0.4450407696 -306 0.4449429961 -307 0.4447875938 +305 0.4450407168 +306 0.4449429433 +307 0.4447876466 308 0.4447108434 309 0.4446476155 -310 0.4445679071 +310 0.4445678542 311 0.4443813398 312 0.4442457457 313 0.4441295373 314 0.4439701205 315 0.4436869418 316 0.4436074446 -317 0.4433313969 -318 0.4431709236 +317 0.4433314497 +318 0.4431709765 319 0.4430649099 -320 0.4429951056 -321 0.4428992073 +320 0.4429950792 +321 0.4428991809 322 0.4428329157 -323 0.4427575123 +323 0.4427574859 324 0.4426940995 -325 0.4426260119 +325 0.4426259855 326 0.4425582941 -327 0.4424612601 -328 0.4422838046 -329 0.4422160868 -330 0.4421166229 -331 0.441977886 -332 0.4418363758 +327 0.4424613393 +328 0.4422837782 +329 0.442216166 +330 0.4421165965 +331 0.4419779124 +332 0.4418364022 333 0.4417645907 334 0.441594768 335 0.4413897658 336 0.4412564167 337 0.4411897553 338 0.4409984605 -339 0.4408126062 -340 0.4407580675 -341 0.440634781 +339 0.4408125798 +340 0.4407580147 +341 0.4406348074 342 0.4405851283 -343 0.4404759981 -344 0.4403500704 -345 0.4402948978 -346 0.4401106547 -347 0.4397904477 -348 0.4397326341 -349 0.4396824532 -350 0.439624032 -351 0.4394675733 +343 0.4404759716 +344 0.440350044 +345 0.4402949506 +346 0.4401106811 +347 0.4397903685 +348 0.4397325812 +349 0.4396825324 +350 0.4396240584 +351 0.4394675469 352 0.4394047943 -353 0.4393420418 -354 0.4392639445 -355 0.4391883562 -356 0.4391398392 -357 0.4389915678 -358 0.4387959944 -359 0.438557371 +353 0.4393420154 +354 0.4392638916 +355 0.4391883298 +356 0.4391398656 +357 0.4389915942 +358 0.4387960208 +359 0.4385573446 360 0.4385087747 361 0.4383729694 -362 0.4383069419 -363 0.4381990265 -364 0.43801238 -365 0.4378701568 +362 0.4383069155 +363 0.4381989737 +364 0.4380124328 +365 0.4378701304 366 0.4378172555 -367 0.4377448894 -368 0.4374820735 +367 0.4377449158 +368 0.4374820207 369 0.437295876 370 0.4372092215 371 0.4370238163 372 0.436981902 -373 0.4368016998 -374 0.4366709917 -375 0.4365565529 -376 0.4363945478 +373 0.4368016734 +374 0.4366709653 +375 0.4365566057 +376 0.4363945214 377 0.4363409599 -378 0.4362627569 -379 0.4361928206 -380 0.4361403683 +378 0.4362627305 +379 0.4361927678 +380 0.4361403419 381 0.4359903539 382 0.4357839519 -383 0.4357237876 +383 0.435723814 384 0.4356793379 -385 0.4355541498 -386 0.4353642019 +385 0.4355540706 +386 0.4353642283 387 0.4352438734 -388 0.4351800644 -389 0.4350318459 -390 0.4348217199 +388 0.4351800116 +389 0.4350318194 +390 0.4348216935 391 0.4347826317 -392 0.4346708867 +392 0.4346708603 393 0.434634598 -394 0.4344987662 -395 0.4343509967 +394 0.4344988191 +395 0.4343509175 396 0.4342753292 -397 0.4341274276 -398 0.4339128118 +397 0.4341275068 +398 0.4339127854 399 0.4338583523 -400 0.4336283917 +400 0.4336283653 401 0.4335961703 402 0.4334266645 403 0.4333842749 -404 0.433300552 -405 0.4332627579 -406 0.4332207644 -407 0.4331742546 -408 0.4331175502 +404 0.4333006312 +405 0.4332627314 +406 0.433220738 +407 0.433174281 +408 0.4331175766 409 0.4330016059 -410 0.4329391703 -411 0.4327436761 +410 0.4329391439 +411 0.4327437025 412 0.4326503132 -413 0.4326064445 -414 0.4325310147 -415 0.4322932893 -416 0.4321530205 -417 0.4320995646 -418 0.4320641475 -419 0.4318961471 -420 0.431762032 -421 0.4316429712 +413 0.4326064973 +414 0.4325309883 +415 0.4322932629 +416 0.4321530469 +417 0.4320995382 +418 0.432064121 +419 0.4318960943 +420 0.4317620056 +421 0.4316429448 422 0.4316029322 -423 0.4314649611 -424 0.4313357849 -425 0.4312889318 +423 0.4314649347 +424 0.4313358113 +425 0.4312889846 426 0.4311176301 -427 0.4310804434 +427 0.4310804698 428 0.4309550439 -429 0.4308267921 +429 0.4308267393 430 0.4306758797 431 0.430564214 -432 0.4305215338 +432 0.4305215074 433 0.4304761861 434 0.4303507603 -435 0.4302473612 +435 0.4302473348 436 0.4302070316 -437 0.4301724332 -438 0.4300275953 +437 0.4301724068 +438 0.4300275424 439 0.4299943702 440 0.4298901524 -441 0.4298619719 -442 0.4297966311 -443 0.4295941908 -444 0.4293884755 -445 0.4292060811 +441 0.4298619455 +442 0.4297966047 +443 0.42959427 +444 0.4293884491 +445 0.4292061075 446 0.4291471318 447 0.4290125149 448 0.4288033134 449 0.4287767175 -450 0.4287522081 +450 0.4287522345 451 0.4287100297 -452 0.4284893658 -453 0.4284410865 +452 0.4284893922 +453 0.4284411129 454 0.4284061448 -455 0.428283677 +455 0.4282836505 456 0.42796347 -457 0.4279105952 +457 0.427910648 458 0.4276986997 -459 0.4275380152 +459 0.4275380416 460 0.4274078618 461 0.4272547308 -462 0.4272262598 +462 0.4272262333 463 0.4271944609 464 0.4271767391 465 0.4271252641 -466 0.4269907529 -467 0.4269561545 +466 0.4269907001 +467 0.4269561016 468 0.4267656783 469 0.4266574461 -470 0.4266171429 -471 0.4264591523 -472 0.426324509 -473 0.4262026222 +470 0.4266170901 +471 0.4264592051 +472 0.4263245354 +473 0.4262025958 474 0.4261693444 -475 0.4259847579 +475 0.4259847315 476 0.4259156403 -477 0.4258481602 -478 0.4258218548 -479 0.4257969493 -480 0.4257487756 -481 0.4256836461 -482 0.4256607742 -483 0.4255501649 -484 0.4255061642 -485 0.4253049916 +477 0.4258481866 +478 0.4258218284 +479 0.4257969757 +480 0.425748802 +481 0.4256836989 +482 0.425660827 +483 0.4255501385 +484 0.4255061906 +485 0.4253049652 486 0.4251170245 487 0.4249721602 -488 0.4249461189 -489 0.4249260201 -490 0.424814645 -491 0.4247909543 +488 0.4249461453 +489 0.4249259937 +490 0.4248146186 +491 0.4247909279 492 0.4247375248 -493 0.4246067376 -494 0.4245252596 -495 0.4243307162 +493 0.4246067904 +494 0.424525286 +495 0.4243306898 496 0.4242090143 -497 0.4241103428 -498 0.4240749785 +497 0.4241103164 +498 0.4240750049 499 0.423876315 -500 0.4236667437 +500 0.4236666909 501 0.4235874314 -502 0.4235353754 -503 0.4234143602 -504 0.4233028793 +502 0.4235354546 +503 0.4234143338 +504 0.4233029057 505 0.423176714 506 0.4231508576 507 0.4229899882 -508 0.4228718782 -509 0.4228359329 -510 0.4227991952 -511 0.4227724144 -512 0.4225436952 -513 0.4224996416 +508 0.4228718518 +509 0.4228359593 +510 0.4227992216 +511 0.422772388 +512 0.4225437216 +513 0.4224996152 514 0.4223835653 -515 0.4223577353 -516 0.4222493446 +515 0.4223577089 +516 0.422249371 517 0.4222281101 518 0.422172198 -519 0.4221387881 +519 0.4221387617 520 0.4221057744 521 0.4219720291 -522 0.4218148572 -523 0.421678867 -524 0.4215360891 +522 0.4218148308 +523 0.4216788934 +524 0.4215360627 525 0.4214808109 -526 0.421386418 -527 0.4212872182 -528 0.4212450663 +526 0.4213864444 +527 0.4212872711 +528 0.4212450927 529 0.421218312 530 0.421077647 -531 0.4209894342 +531 0.420989487 532 0.4209177812 -533 0.4208865898 -534 0.4208590167 -535 0.4207220229 -536 0.4205735666 +533 0.4208866426 +534 0.4208589903 +535 0.42072197 +536 0.4205736194 537 0.4205432204 -538 0.4205065883 -539 0.4204169494 +538 0.4205065355 +539 0.4204168966 540 0.4203891914 -541 0.4202709494 -542 0.420224017 +541 0.420270923 +542 0.4202239642 543 0.4201979758 544 0.4201635359 -545 0.4201325029 +545 0.4201325293 546 0.4201024736 -547 0.4199849975 -548 0.4198871712 -549 0.4197798368 +547 0.4199850503 +548 0.4198871183 +549 0.4197798104 550 0.4196252004 -551 0.4195158061 -552 0.4193934703 -553 0.4192237533 -554 0.4191110047 +551 0.4195157797 +552 0.4193934175 +553 0.4192236476 +554 0.4191110311 555 0.419012386 556 0.4189145333 -557 0.4188031845 -558 0.4185742011 -559 0.4185236769 +557 0.4188031317 +558 0.4185741747 +559 0.4185236241 560 0.4184201458 561 0.4183944214 -562 0.4183447688 -563 0.4182207955 -564 0.4180463509 -565 0.41787703 -566 0.4177670546 +562 0.4183447159 +563 0.4182208219 +564 0.4180463773 +565 0.4178770828 +566 0.4177670017 567 0.4177384779 568 0.4175663046 569 0.4174468476 -570 0.4172682036 -571 0.4171848769 +570 0.41726823 +571 0.4171849561 572 0.417080395 573 0.4170362886 -574 0.4169296674 +574 0.416929641 575 0.4168813353 -576 0.4168014156 -577 0.4166510314 -578 0.4165654861 -579 0.4164509152 -580 0.4163564959 +576 0.416801442 +577 0.4166510049 +578 0.4165655125 +579 0.4164508624 +580 0.4163564431 581 0.4163250932 582 0.4161199854 583 0.4160572857 -584 0.4160180917 -585 0.4159334181 -586 0.4157927531 -587 0.4156931836 -588 0.4154709615 -589 0.415443811 -590 0.4153113862 -591 0.4152067722 -592 0.4151275921 -593 0.4149858178 -594 0.4149130555 +584 0.4160180125 +585 0.4159334445 +586 0.4157927795 +587 0.41569321 +588 0.4154710407 +589 0.4154438638 +590 0.4153113598 +591 0.4152067194 +592 0.4151275657 +593 0.4149858442 +594 0.4149130819 595 0.4146729531 -596 0.4144590504 +596 0.4144589976 597 0.4143572096 -598 0.4143302704 -599 0.4142094401 -600 0.4140924129 -601 0.4139592751 -602 0.4138737827 +598 0.4143302968 +599 0.4142094665 +600 0.4140924393 +601 0.4139592487 +602 0.4138737299 603 0.4138459455 -604 0.4136362686 -605 0.4136080088 -606 0.4135243916 +604 0.4136362157 +605 0.4136080352 +606 0.413524418 607 0.4133501054 608 0.4132487135 -609 0.4131160511 +609 0.4131160247 610 0.4130051249 -611 0.4129740391 +611 0.4129740127 612 0.4128680782 -613 0.412837415 +613 0.4128373622 614 0.4127119892 615 0.4126791075 -616 0.4125751802 -617 0.4124947851 -618 0.4123330706 +616 0.4125751274 +617 0.4124947587 +618 0.4123330442 619 0.4123064483 -620 0.4121374971 -621 0.4120678249 +620 0.41213755 +621 0.4120677985 622 0.4119360076 -623 0.4118715384 +623 0.4118715648 624 0.4117428904 -625 0.4116109146 +625 0.411610941 626 0.4115017844 627 0.411481131 628 0.4113286075 -629 0.4112463108 -630 0.4110173538 +629 0.4112462844 +630 0.4110174331 631 0.4108394229 -632 0.4105907634 -633 0.4105139073 -634 0.4101629052 +632 0.4105907898 +633 0.4105138809 +634 0.4101628787 635 0.4099986023 636 0.4098852727 -637 0.4097632803 -638 0.4096657973 -639 0.4095620549 -640 0.409416187 -641 0.4093790267 +637 0.4097632539 +638 0.4096657445 +639 0.4095621077 +640 0.4094162134 +641 0.4093790531 642 0.4092723527 -643 0.4091630639 -644 0.4090574199 -645 0.4089596992 -646 0.4088231016 -647 0.4088041913 -648 0.4086547843 -649 0.4084877347 +643 0.4091630375 +644 0.4090574464 +645 0.4089596728 +646 0.4088230751 +647 0.4088042177 +648 0.4086547578 +649 0.4084876819 650 0.4084118559 -651 0.4082863508 +651 0.4082863772 652 0.4082518316 653 0.4080320657 -654 0.4078376807 -655 0.4076548902 -656 0.407535354 -657 0.4074433909 +654 0.4078377072 +655 0.4076549166 +656 0.4075353804 +657 0.4074434437 658 0.4072420335 -659 0.4071415924 +659 0.4071416188 660 0.407027312 -661 0.4069112621 +661 0.4069112357 662 0.406769752 -663 0.4065997444 -664 0.40648932 -665 0.4062134835 -666 0.4061617443 +663 0.4065997972 +664 0.4064893464 +665 0.4062135099 +666 0.4061617972 667 0.4060348659 -668 0.4060129712 +668 0.4060129448 669 0.4059170464 -670 0.4058050638 -671 0.4056596976 -672 0.4055702172 +670 0.4058050374 +671 0.4056596712 +672 0.4055701908 673 0.4054526882 -674 0.4053703651 +674 0.4053703916 675 0.4052619744 -676 0.4051706716 -677 0.4050839379 +676 0.4051707244 +677 0.4050839114 678 0.4050310102 -679 0.4049275583 -680 0.4048025022 -681 0.4047124407 +679 0.4049275319 +680 0.4048024494 +681 0.4047124671 682 0.4045690818 -683 0.4044315333 +683 0.4044315598 684 0.4042882801 -685 0.404154482 -686 0.4039425865 -687 0.4038427529 -688 0.4037263068 +685 0.4041545084 +686 0.4039426129 +687 0.4038427265 +688 0.4037263332 689 0.4036760467 690 0.403583212 691 0.4034757985 -692 0.4033589826 -693 0.403220589 -694 0.4030926277 +692 0.4033590355 +693 0.4032205626 +694 0.4030926541 695 0.4029039211 696 0.4028226545 -697 0.402755148 -698 0.4025953614 -699 0.402450814 +697 0.4027551215 +698 0.4025954142 +699 0.4024508404 700 0.4023101754 -701 0.4022780597 -702 0.402171729 +701 0.4022780332 +702 0.4021717026 703 0.4020247253 704 0.4019478165 -705 0.4017364493 -706 0.4016885397 -707 0.4015977387 -708 0.401570641 -709 0.4015012065 -710 0.4013771276 +705 0.4017364757 +706 0.4016885926 +707 0.4015977123 +708 0.4015706146 +709 0.4015012329 +710 0.4013771805 711 0.4013263657 -712 0.4012401074 -713 0.4011601349 -714 0.4010814037 -715 0.400944938 -716 0.4008109286 -717 0.4006492141 +712 0.4012401338 +713 0.401160082 +714 0.4010814301 +715 0.4009449116 +716 0.4008109022 +717 0.4006492669 718 0.400531421 -719 0.4003784749 -720 0.400226374 +719 0.4003784485 +720 0.4002264268 721 0.4001478805 -722 0.4000406254 +722 0.4000405726 723 0.3999716927 -724 0.3998336424 -725 0.3997517683 +724 0.3998335896 +725 0.3997517947 726 0.3997031721 -727 0.3994946836 -728 0.3994120436 -729 0.3992452581 +727 0.3994946572 +728 0.3994120964 +729 0.3992452317 730 0.3991151311 -731 0.3989169694 -732 0.3988042473 -733 0.3987856803 +731 0.398916943 +732 0.3988042209 +733 0.3987857067 734 0.398664269 735 0.3985673142 -736 0.3984359195 -737 0.3983484991 -738 0.3983241217 +736 0.3984358931 +737 0.3983485783 +738 0.3983240953 739 0.3982504614 -740 0.39816246 -741 0.3980996282 -742 0.3980129473 +740 0.3981624864 +741 0.3980996546 +742 0.3980128945 743 0.3979900754 744 0.3978655475 -745 0.3977282896 -746 0.3976407635 -747 0.3974670848 -748 0.3973722165 -749 0.3972935645 +745 0.3977282631 +746 0.3976407107 +747 0.3974671112 +748 0.3973722429 +749 0.3972935381 750 0.3971188557 751 0.3969426415 -752 0.3967768069 -753 0.396643088 -754 0.3965085768 +752 0.3967767541 +753 0.3966430616 +754 0.3965085503 755 0.3964347052 756 0.3963336303 -757 0.3962246849 +757 0.3962247377 758 0.3961006589 -759 0.3959192945 -760 0.3958587605 -761 0.3958067044 +759 0.3959193209 +760 0.3958587869 +761 0.395806678 762 0.3956605196 -763 0.3956133759 -764 0.3955843238 -765 0.3954913571 +763 0.3956133231 +764 0.3955843502 +765 0.3954913307 766 0.3954052044 -767 0.3952515981 -768 0.3951503911 +767 0.3952516509 +768 0.3951503647 769 0.3950491577 -770 0.3949205362 -771 0.3948132283 -772 0.3946939298 -773 0.3946074866 -774 0.3945200398 +770 0.394920589 +771 0.3948132547 +772 0.3946939034 +773 0.3946075394 +774 0.3945199869 775 0.3944517409 -776 0.394347708 +776 0.3943477608 777 0.39426314 778 0.3941691696 -779 0.3941472485 +779 0.3941472749 780 0.3940587452 -781 0.3939841342 -782 0.3938982192 +781 0.3939841078 +782 0.3938981928 783 0.3937222427 -784 0.3936012803 -785 0.3935053556 -786 0.393446327 -787 0.3932816016 +784 0.3936012539 +785 0.3935053292 +786 0.3934463798 +787 0.393281628 diff --git a/main/train/catboost_info/test/events.out.tfevents b/main/train/catboost_info/test/events.out.tfevents index c28526daf1c212cdc4584e545760c8d6bb832381..128a556e2d2f0e46d4a0d7368468cd27c3bddc8f 100644 GIT binary patch literal 43210 zcmZ|YcU(?={6FvuSy@R@+*bBV*|&MwWUt%2&FsB57YT_dvNDoUwuDmAg^+Bbq>xc5 z%E}(U-?z`_`|WjJ*XO?fyZ?ATUhnt&obx%KbIygzI`LndT83@~bvk3l!6NakYucKZ zGSA<2>exwBXU?=vS?bWL>Hq#;&GvTwRi&BPBCdJ-BWywX)EvF`<=nvi7f@m0zujYn5kH zP~MII%up*!t+HNODJSdQe#dQURYI$(n}c%vdB=y8QOwXvt0pb7%gL(!b<|e1vev5F z@fCAag+*P*s#Qs?8rcn0`ox9{)vAMQZ z)vBd8%jaY@9qUq5tx9XviI$*7IBThaYE|z+pzK$_eWq3pT9xn@ zs5zcJcdM17Rt3eC%&D-o^+-O-vch^=HDLp&*ujTask7>9m1A8{wMN_9kur#BG|(!` z6ziM{f5j$LRI7$sHF!U$CYRS9QmaN8p_Qw0|i} z_aXaQZmw1Hw}E=}Gvtn1wa}^s9YOhp9yi&Ui>#Jf74_RPr^4YqTu-U9T4~jYbD%c< z*~-r;dXc5qD*L&hj@(P<8A?S@Z55h;iZP4hvsbD%+N|0c#dG?4l73++^(9qXt#UaE zDzdyS?`@=Nr&Wn=p!PVLr6ix)oKrB)rZ>U1SgH)>d#?3P7VN3ELl z%pzwRU9PvRug>bERlg!Y&EH!uPOUm?)uf%EUT1~WC1n&9I%(CYDWJSxCh~qmsxDe( z=mx6Y%j%cZSzWd2K?zXj`z}sUt8Q9#BOj&=?QA$g17&X)my7{jX_lis?v=r)Qi6Q zXjN`AP(3nM@lH>wzFM{HPSKnhZdovL>mshG^CI+n|mv(H|pa6j}dj)$s$MKG%HTNv(!zRm+v2B79dn zs?{*9x;zt9?!_x=tJQF=O6&$|QqwJZwHl#SotuDK5X+6(_7 zr5Do}t5rL;g8DnM*?qMdr&a#TK_$2CRa&jaYgNJ^P~O)Xo2k_Vt-9V8RLPMo3z0I2 zz9wo_f+MI^^D6ObBh@6Wn)nx}mdAV@)LE0YYDaNUKZaF}P^&3g)jqRO&bd){$OL{u zlod|Zs)~uAEX(gOr!1qG#x$+!7ze6%n%Pvfnyyuiu7MisGT-F=yvUlNRmQWRqCGEF zQD@E6s=)!Ez7+~?rdG4Gs?=&wkp((>lG0^muK=^Ps@4Kfr3*Q{RjWB#)o3KB@wVgm z?3Gz_waU2{sI`u_E!0`_w5m@VP*W=%sG(N#wdz`VQ0FG5MUv8sX)MqxzkHzV-p}BB zrBn;GYS;(!oU^x4i8_zfS&OtPBo$QSx9t+uYOz+GzXhs{)vng045F_kTIFyNRGSco zpK7&Kt3pnL3i?`dNHzH-D=aCn!~g%1C9@3lDQn3dV3$W64pFmZ#5x&(tqc2cRn3+Y z3mOlslFd5p z(RlaCSgmI6#I`;K)}&%mgqp1-HsT?$w@X9PiRr~i))5Q70BqKZIcL?(gV@`{z-oTY zld5LxiDmi$8|5@SR?RjLo9_uszq;2aVg}KhC$ai#fUPswIH{Qzv3s+C<@Y`4uVx#G z&6)_TdLj33YPN}34`*PPEi2wpGjC$)1A%#k)#^peC`PiGn0*&uc^vjOP_r$>4z>hV z{?x@5YPOZwbq8QASN83zX4{CRRRb3AA#t*rZ70^X1~5aetIvt)zGN>7KE$3^26le( znJhKiK`gy2uv4u{bXK#S#7-9lW*FLYg_`XmHnJG7gN3iBso8E~+x`^DS%Q-LuD?l4 zFGk`^Y)>voyBJ$}xSH)Db{^BbTi-maW`4w$q(hO*QWsgN*9s!Fu7WIgjLG-qd z*ypRjJ|(|)SF`=ZW}X7pIkoOaH9J7;cmS~Q;8Fe6%%9kdoxs9cjEGjV0Af8h0y|Xg z?Mh-sF_MGCtlWTA*7XWfvqQu>F9zn_CYPO>9VYf>4zTNywR@`B5n?T;0P8n7)k4jV z5(^j(Y-)J3V*SA9ax@iDfkh=9RGiJ~6!*Nf5D1HG!4gU)({>P7(W35m?lpr^VImG_iwa zfmQBZe59J4AvUWhu(bDG%M&w*-hzqQm;t+&>-AVQ3nBLQXa1b!HJ8hh7ixBvSZE$d z8=R8MFA@py`qqs8HtUN5c7_Jw8tU8 zs;b#VVvWLqoe#->Ud=8MYl>-KGUxJRpDZ$x*w_$AtJn0%Ka{2yy!!pJ0}4vk*7Fc zIFj8bw%r+6;n0|YRHW{E_M#9&%w{;SPHmlPtJwo$Rs(_cH(ze1W)F!K>jUgBr=q;Y zmPN)A3+o3=XYb8>WxePvj##&zz=G#C&qGB@_K4V7Ct&HN{v1)Wcw)!&z)HtiPf@eS z#FASA8xX$VWC;?zB@mn40NBvXA$+4U2$o1JsVcAy1*k>+ZaL~H`4-Do-PD>0)e@;R}Hl8|=#QDCN;y&yKe z7_e`}|H)9Zm&BY40E-@K<*jC~h`q`UY=Um(Gh(_Q*;jWmvFOkFa&~wP+Mc_oW+}v; zegO7Ql#QdBr4qaK3D~BhvHwxzYhrd-WT7oD&#BYWh_(9&Y0XM+-A+s|Mv_i!Z5psP z;tg^k0I@TxplYI>?1LUhrrrf{E8=L z6uo^SRyht>>WkrR)hvtHgWJGt8!lR=W}k_5x&^Gzi7RK->@o_JdeZ7_c#a8;h!o{7Gzd zFtEzjxhtsIFJeQE1GAa#%7^tPPX-R;0i*=;$kW-i2rYymbm)oAi!U!<)d_HHAv zvn`GcSEsoWv-AeGU01TCnyn;ubuF+K-qHL?(kO~tMJ!+qu((aD_${_%tBJX|0Bg6o z4(}%=b0gM$5wJ-+%GgqGlC2>&a4xXlzqaxdm1OS37EA$FyI>uYMd4TWqOg`&*|EUl zD}CjW=mc9wEDFztTeSwj;W`~*Ox z?I(822H45tAF|YG2Z$w=1m-*?ocEKmH-BO)t$__qZea4+P81nHY^D{k;+41Y$z>4i zATdu&TiwFK#I? z#3~j7_PuHVZ+d0gF=DyQfz9swmG3x`9Vcdktp2@rHPuC)Aa)@Sq;*;t%{v9%pX@~; zkl2u1!2DMo}O#0J6Z2nr-cv;cn@sX=!2!z>@2a8 z*xQ+2k6Wo(D6zgDA+1$VZT@a*5F-gAHU^7)=<>~ELoZl3v4zN%-7T4)ij-;Rh>iXL zMHYQH^Di|!Ppm;Eu-HDWniDgMA|r@dVA`-wA9&@G%t*`?)3%Px;I}i9T_83b4}gDK z)qLtAFA}@)4tiU0?JwS9%d|_xnqzORF4niEG@b6y|NLzrk;F#4g|z7+9!*=N-Q=F(o!-btEtnXh;4Wc>}rF^X~gtmB-e;pA+w!c zVe zHM>RZP6DuS|DXNT>^8Bs4}k63oA6uB?htz%1FW$3^o?qEm)IR-lWtsZNz5ola*x>f zJCL@;wH|-8knBFOOSge7U$BdRR4rKyvE8?T#mC(CRTueySY2cuHS{JwZxFpbB<76l z{oXGo*ITaaMIn~hm>W=JfA_Q>RHRO@IAZ;-1G9UQG*`_Y5pzIh7C+ij&EkpO!y@Y! zyf>McUKIJ5Sdp7hT6Sf>ZCO?C<*?K!am$n+g<{zuvi zVjGc7wM^viII_2w#Ks_7Ir^{-^=1@Bz9QD~3iM_bRQ|l0B@;`=wDBM36;QJjV(XF3 zUfPE@hO)?1VwsUp9x?(D1kHL-G-Hff%{8x<*88nJj}`^;=j9@j-$IrC-EB$y(lt+*fC^lnUA(k7jfcr<&Om6w2 zHv|3kibJutc9Z=NP?}z_WyF?Yk^MGw@>8?r#O#o5dQ-NUnz<00hZoMggA=!^*$QH2 zcq(VP%;qOygXqnb*hf4k50!M{<0jckVlS|_b!{W~ERk##u?l!9kN)A8M7>G2nph`1 zxlXQJXYvL^^yWsaH?k$O|rGbl93fm zv0STW>xj)k7FaB+zM6Run|TPX?&-VTM-$VTWiJZriRD8!|DrvA0F`V5vCqg>`3~hG zCG#Yf|1cC8KDtbAb&+1gHe%YI#ZSr;(~I6V5}SPh(gqaBs-tF`h>Z#WR@pFki<)^8 zv&7!qwq?9mv(3al9E7y6;f+jIF45Z-VqGxJKVxw!r5Oa^JBTel29s>py&9((#YlD%Yk_P=#=C`7 zq-48@&BX0Ya7G)x-Xz;iYyhrYuAx)Ps?&UlRmGJnz<-Fhn(ZN`*?KaJ+=fivs_8|M z2Z+tV`&*?oMR|)YnLn{OOiTU8dIl9KSpcz6oaBunb{T4Rkk~t%WIMlGDQb3z*mOLV z|GoBRBQb**$zft$ku~hwkUvREc7)irqj0@_a?9jfhh#^IeLW7$>~MIRy2xY1rr~XI zbKmy-O<1NKC+3R>;MHl=qJC5(B;d%x=#_yhbz~>;qB{Enq;Sm1tS}Oc+n;`J437}PVz!} z^<`=nOe_R1)BVx*d(|w2m?g5BkN@ylq8B4MORPSgB~?$|@SrrwLW!B}g>SU4Uh&96F_M(1jZb~zVk(?)X6VocxZDv%n2x6OY zEtx%Z)Dbl^60^V}t=gA=p=K9|S>rm{uuSY!HM>acF*1*?cTK(rh>=_())U#Bxsh3v zW)v)v*!d%{;I-KO$K>%ru*<}@;V#J0>0J|b+7)6KarYYIocD~HT_qNR8^fDZTkRyK z%b&fLL=mfk8^bT*=XkFy*)?KT{xFj0Ol#isN)}DbXFss*G3EG)O0w(3(r_g16N~Wc zXuas|2C-gPg}MF0n(%UM^YwH>DZGB<~TcfXD6O>}9jm>^`xsc-%T>hRs*A7-Eg_v2TR) zATKq0KZR*OCNc+mX%6m1IXn z>I6$9RuPN*czAR@HG4vAAF{@!HjGfSr^Mp6LvJqS0{I)2Eb=UN>G;PmM!jdHuYl~@zI?Uqxn`9}(3|`P%>AN%d!?k3o#7=GmHpBH9zZ{V4HLO^a1Rm}iHnaH$;|5XWS7H}&lGTcC;j=`tZ^Q;-kv&{H z2GRkL>^rfK$S$-S!&iW0KZxn^QN$e~>hn}70O-|yon9Q2k5H7z4%5Y*=AyG@oBNAp{B_OL!@mX zRv({)w{LA-KwackVw17ANe+2UHuNHG8!^oT^orv?{5o1M*mh#ia3l+UcOOedO6Egs zEWUZFIlcLv`9tVz==XQkf(hlgD+@+b&`+Fl~^2Ksib?2)3Kp zX8dT;q=eGia0=Yaw+_kth*{%E-W_eu_cx>HZ7;Ds zcq+S8JZG}O5Nsc@iO5>oyS1f}$h7^$HsHxMr@cxu^Y&m#wI3Gnyzs6)$LEL zIgVt)XLBx6vH)V&F)eV`8GeN%*+F9QIFhH`V)-nQ>=3bBSmd-@lYG^^9VX_1Y3oau zvLdDzBRN8>8}?Qz=x~~v9VPYx*|2soacXvqSS7sUY(3m@kD47PRtuT4Rm6T`2GQFI zVpjNi?7^ZU|EgIau?x7on!7siv#CrwNh}v$rnSbkJg!a)B32TMT-SF$@3~~!DPr&O z*0J!=N?u-#VkD=DO~6T}jI`fJMM`#tSP^7XhR$uNX2HacATu1k@gGHo5OYPgeC$kq zKdCE{eRZEDW`UFR3(srvI|ac)iPgiy-Sv}OZR$;?g%SIMhr8XCvwzeqoLD+u-NW3w z#j4pkV)^k#b*YIRk3=teJ5S67i!8EaOm9k)EP|L9rukp7p9aH13nne=R;TPyH4N4}f*=1ta zk##TO%jcvl@(QteXJBW3qq{%_y=QF&|7T=+=9On%yMU8GG|f z8^KqAOuI#_PZ&({(}I1I)M>Yg{W%LP&j$~GHM>JBHy$_tqQ&MB(~FVZCFXt((h>`u zovmi~h?PNB&!$xoHM>u&)pr-NlNyB*!*xvTkz(6RVvaT zM)Hu@dYojvwb6g8SuC+gJZ>%Z3GQkZN6Z^fuE6cd5o-2`ST9WT3MhJmm{If=Ps|&8 zD`jR`LCqc$TMz=1Y*2nI9{`z_K&%I{6}4LPXL89Bi7i34|9F#*>LQ;I+lg%Ls5v|m zU9s#%;VH3OcmOIq{Cfa;h+nqiRvWqz1k-C`tfiRHq>oe^}6CutNcgIHZGa#XZW7!@g5Cb2)rg3Jx}YW9xU zS1d9rPZ+<{(OG0)-S3IHB0KAl#4j2p`#@|vUfpN6@8^AqWFLt|;Sv9Wg5N1f_JvqH?v*bTtua!a_LW#H z?#y?NOyuQNrhOw;99h$+`QKBTLG<>Wm=RfYSzA82B>O>ZF!pw()cAJlw4cNlAe(<_ zGd~H-v|q#qA+xI*%eM~6eiO@sy*0@QF;zSh)l z9BT6TAkvl*YlCcXqw+P>Y0HT{$H%^ijA514%!Sxce8@amVPIWiy5iZ3!U|%o@%30h zd{_QTE14^?QTU!~+}2+U)M+b;-NqvOUmG2%W~+#0;7AVF@CsM6)x>Hc%V+7%=cHbY z#En=MzKUp?)Z!GSNw$XAPE7O4>+Gs#?!?^j2dvdy>ziyaM3HNWokq64;hLkAW)N&0 zv5i<{%urh!HS-|07*FMRr*^Z{Y(25_c$v06Hkdy)$s#up8;_Ui(ua*Isna}(+2haC z>+jmezf&-Zk$4ev$6LpoZaqw%@&(&StR-$kJ9XX7UyWqiCSn$N%m3%!Xv*;Zoh@s_{I@6s+c+eYkf ze1l;(ccS^eGD)$9PV zqxc6jI$cUCt!DnjP9gJXJ$Z(j1rXDGN_y=av0_-H zU6pnGY%1AdVv8{?Ew%>VaU?rJ>=o`ymibs)s*5~IO!M6|{$v-E&vv4>W5hb+zT{G5 zGaiXvu;auQB5Rp8jbD#Rc7j+qj$}xdkrqhoK7O?5w*CA$8i`ChNz5A4Y@JT? zW1m5!1ramH-a0sJ;*FtXr-%jP=BdEibbdCK>@=|#c(||LeZ-qy$<7e7#UfX1vEcor zWWmIyBTH!ZlGhTW7)c1Rc%0;~m;%G;07!P0SXEr`F3cX(UCly?8L`O7HD~gvSr{?R z&l_CLZMaCCRrb{#PHZ4<=nGfyGr5}zc8=H_WR6eEn^BQ6?L4ti{7QS~NW&j$7C}sh zEHEr(nVK1i?Zhvc{o0i_+0cvLE)diF!YXvKj*HX_c9GahJeBjAH}s?;CA&n-6DOI= zCw-8bMG~8Zi$aYnNselEnb>J8@~KVSAz}v6+ZAFDkQLY&8l`4ei48zzF>m)!HH#uP z5s%woeYFN^c8!?kRNk}al#iN46VvSQCSTjezaBG+kz6OX7q_8zrVrzdp=39Rx!{|p zdY!7oP?3_|B=!(*!uF5)lvA@?#B4Baao*yqi0Q1e7lqrzM&g6m>tVx8Zo-1yAr^pX z_a;Vv)1GfDsAdVo#^8nXedH+qt(s93nMiCg z_BQh9@WPZP*%M;BkX;{Nj;}Y#o)Q~`Mf!A(Gg*Q}kUCHc4;W;rM?5)whVAEwaC1NS?!=AW!g((uaJ4T^-fi@SHyPV67=!S zxh2H(qR3=o%aJW9k@upSr4Td2b8_~dF*VgJl~`x&Evb!5h?>17rul)-;5{cMs974Z z;n-X6BJI}_Gl-F-6WfOiUS#T+XKMC_SW*0H6yMgPf||W0=7`MJ%~P*t8N}A(o2Qi6 zs~&2WNo+c@Z(XaJJO_x8ydzc&zhoM}l)6M|M#0__dxW1P2R(FntY#mGMPk}O{UQF` zD$_m^n~1+J@qBxRA2M~Nvajw>#7afL&l}QRlT8){!Lo>1;a>nc+xZ$ltjM&_!~!tw zoBpjEjYP69#0KI^)Te{G%u=(j#7g1;n6P8fM`C)>+c#oPSY$%HS;$?cO>ME0`e4Fe=VHGi7 zJWHI7R&%mV*<-{J$~_p6y3v3nO`3Ca`f(pSya5PO7a@1A@zSxZEb?!>BK zT0QeBCX0e#Yl$to3PrA-sM&gA&5+%V zSZ?w?Koq%wnA2rQ^E&#B|2T+2Fi&C)uK;VYf7E9xQZg@MOEJwZ`ASPQ+ej<{nbn-P G3I7ilfajk8 literal 43210 zcmZ|YXIvCX+b;0YRYcZ+0mYorHOrbY%wk?;%{gLN*MtETv!e(KCJ<197%_pFV3si} zDCP_zpkhE#1i^&oJye~;)z{SfeA?&3|99PWcXxGlbr06oiT~L*FgO(bpZ|2cRqV-M z^($DEw>xmp1L#_W%7q^Mzmg+DPNvJ4N^Zjb#C?vR~dLU#$uZSf*A5 zwJLRM<9zkL%kdRz^_x~bk7$&yZm+zUu2zM#s&zuceAUCbO%f@6UN0;QYt`!3pmOTj zmRBn?t+JcgAU~_b)vBRtRYa?T_Ser>CkEDkr&i`#^)}&;eAT1#)>otqVq8VFDyB3j zvpqJEYGt8St>)Lu&w4*)J9k57{jOEJAJolPW*3Xtsk4e{)r*dxiaFhNRjcA!Wpm0V zKdWouHlAu#LaQo$uamEWXE}@{WfVOusa28FK@~1n(Lt>&wd&uKwez#q9C3J~R;9FR zLt9X~ZJP_LRcWoVa<7%2)h5>3MXkzcRXqz(In8qxlhTd-zq5vAS*@D7sAhiFE32&$ zYGtKW4WHD=S8ul;yQEg-v}&vas2UsY@U<)}EU#6A_EgW$O7rg2TAgLBRnwb-dN$4X z7%9EzMg^@}cB5K;R%Nd%gVm~{RvqpDs@S^d2WnMGtD1&a&Cl}k9FnY7m9?sIUr?K3 zuUV*76|K5!tdgH)ex+YaQU=kDs#@jV7S!mcc1~(lO{?}MR?g4TeRyzHt*UF)D@Rb3 z3qD9us~TEWsT8PrUr&3HGKz84)T&NCmGUbb@T3TzQBu{?s?+U3MSOk0v#333I#yj%E;l)OI1&+ z9Q`ZgS2${Ti$zqSRDWpIh7O<{bAnA~08wFmtt#`zIzP)izT!c3Rs*ey-2|#=yOz&M z=|xsUt@`h8P!}uCDWg`6NLhe#TksOdu3>^&wbH6q*=6%9JlngL zvs$&*s$;uBS!Bi+RI4^xmFNWObwNiVtbBWs?&p;jHV>S}K3{0g_O^fj3Q#JDhbpZ9{ZNF5t z>ZMip%Ya(8XzFoNx(RvLvV&G#Hv_e*W8ePpF|OfS<@mFBe&-Jq zn9cheQjO56Z#O}GFrNFa&Kjvz2X}+Iaj|W%T8+}G4(^}~tpmEK)o87{i*4}0mf)mf&-{tWABAgmQ;>f)uT12s)?=Q)M~6&{VD~@%Rba(_drxQPOIAJK>c2^ z`aR0h3pHM=vc43{KgtD`{kB`JCTLaq6Hq;GyqT(2|7z9ONKi{(`DCiqM6KHF532sv zPW{yCKdsuc9F$e}Has61L^mdBRs1ke8D_fblqJ<Q2)PFmF^{rCKH!x+BdrCLqbsu~HPM$Vhv zP_1TYRo6$LF1bbVIWH@msZ}knf%08hif3%8W@(l21gL7J%wea>zWjnyXbU$AVg5Red`ty%?8^R*mlqD&OSS5FBq(F1)3v(t zP1dQTBVGUJO_ot~!$7aHmTm|3>A#`Z)odBDt_EORUtYSTX3L3%PXhM2ar~cZ=1OeL z5MZl*PF|~KD~MIG2UdT{R##%W|MD(qH)3Pj0&C^a@06Ok6VufNW<96tWHnn!Y-wp= z-|jt%QnOXW$`l2bZILxY%{++t>44=PI{7y-y=aLivC;2}=3g`+-=k99e zMQpw|u;~#K#;KV%Fd6R-avFlBN z{pHiUfSLsmds`RSg74whY8FVW*B`)+rXAj@WjA8F zLIVq8y2*K0cPO!*i-FBaUspxV!ic5K1?JrR&RaFxPprKYu-idn&ZyY|VtdB`GhBVq zRLu?&t3CqQobN+M5Yvm693qy`3)l_U{&&^vFtN3LfbDh|&Xc)JJ3`D(5A5px7BAFk zM~P*%19sl^RCQtoF}84G7wQ8mpPY74%_4|p*8n!Qb9<9}w@8a5wyz4X5>{zP)oD@0 z=9B;yQf_irVn&g6jF@#HV9O`pZ>VO+iM{_(B!7DC8{dzMlxZi3o%jvX%GY%JOPzL- zm~9rMIsSETlbW3(cKaQ$89gR`C8nE_cXgj8HZBF&!?OXU)$9zhzwZN!YwO-q&CU|@ zj|Da=r4Wxz78y+};3_bW)3x{+M=~R^|1hn2=GK=~q+Yb-95KIWNNcd+_B1s+Ppsu> zVC&ZZnXP6Qh}mJYPjF@qS}C1NMSfJHc+;dxTB%f!t005c!6 zDN&tvg;<%r!1jHf*HO)`61xxxEZ*Tpu&PVBTRu%1@WGSut_u_}v!ojqc)h?s6_-qjsP?A;t-Mc+7nRz{XjZ{i+mi|yl z&0Y{2Zw{>JsQqJ!=|zz*iM=cc%rI@&el>eVEar<@{tC~sexS)^D$-JkC1wE22(c@p zPJ2x(E)&?{CO`g{vArQy6N_vTb(1%B45G-l#5#Y5wEeHLmQ#_Ey(6~fEwCSso~_g@ zjoAHUU@mjRC#%_eV!lbhUWcrCM$9P2_JLTXSHRYk{jpHZ(uwtb4QxbnTRv_w?IW>` z&w*JN>ij^R_KDa#WU(J(7N}VUG4rR8=F%~A4l$im-qrn?Skni<>Qo=b%TUQOiPe4t zZ2P3Cyy%rIi`avkzJxfSqsjjc-(XF}7@CHE#hcG}rG0rAd}U%=#*@ zetX;SIVss!VyiC#D{lAgf;#ORvHvarI~(S+NzJ|!3qJ)cs`MIuYh(~D`9W;LF<^V1 zu9!<{lKmt$;t;T)Zofj+ESK1hBfv)Xdsko0ei7Th57?+%Q_|GTKtBrWw+Gm9>z=)c z8AVH$G1~?#e%#rDYPOu%?Lc7Ndh9gWp%-bc#71lawrA;@-_&U{3^t8!^BhoyH4Rrygn{$N6rWb5Au?0@RY-d^VOAE=?5UV;7*v5n%e4~!HO6|0I&CYlPtAapwXk@rX4{A@Y6xsm??Y3F8AMCA6YEhQ*lzb# zgVbyXv03$ib&jgRhg+uYBzB_)us)5$9;nkoi1}0kHsN#UQEIk}Se!Mms-cq&#EhaP zyNM;019om$n8{YVV0(zUTLC*&uvVfvZ7(tBvcM+xy~Wp?EOH+)Z%liZJ!C4S>1O6# z-J!(nOGDc2ekb{Lt7Kus-jxFOadhQ$b=rPn>xu*Guxm4a%_7qd5Ucq+uw&k>8>-U| z5-VT9miZIt(H$f>=RhKm0xBQ<`Lv#5}SfZPv~Ed~!(^ zMQna1ut1L{C)8=jh*kLlY-_3D#%gw)*xpaTY}1da;a)MZS99#B%GjmFl>?E-P zpCRo@3A^EHc8b`)SY-91hi9qTX<`eJW$)~HfS7Jp-lTAb*n|uya`VB1b=B-Fu|Ly+ zb?@!-Sk0n|Sz_Aw8;3L1%t*`)({_gM=i8zj+c{#R@c@|jx7|Q#deM^e#IApYv9(OP zWb$R4U>AtB!?8_!`+?tx$+U~a#=eKN<+Tr4sEfQrZ2Sjc$E*D`nYl!fmx&d_wD4UQ zYfzd&uq(tKy@fQpejf z;!R*RYzDnjvwOt0!~wH3JKcboUX1NNu_nm&)voHNW)Fx>Kz1~y(tm39kl4T1p~$NC zxA=OKMLr@nFcw(P^e;wr+GAqPka>)En@h|fTJnV0Z7i}yN@fK$drGYM4JdNNrB5r> zES{Jf7J0Q!&6aAGK)g!+YWAF1 znJbX?v1mtrMIzIZh&8_itcLl`^6Io?V&=&H*py;2Da_8B6jF%!AdB*yT7=Scg1sR2 zFS6*ws{BMH*-K*GF2mTypKjx?PJ2Zx71Krz4=~yI5JjdETaT>pyu~J4e}cUx_VFSV z+0V0bCn{1e*c)P1F>U7PN<2?W_LkTaWC5}Dc+Dl*J7O)dlWsY)_yJn7G-AOxw&?JI zJWm?L*xnO+ZiJR3Jt%X9S|Zs8Vuz5GeDv8~&C-cYItOVzv-{Lhvya5ao&xq);7*fC zL5%Gau}Q~)U243K(~N>;5c?PbtW}vKF;t{vpNYLW3T*G)QKi%@li0!|z#0~R&*!8} z%OdvtFtF-%gLzj?Hz#jW_(E(l7WwY5;S?1qSvIj9QIPi0rAZYv%OUm?JGpz^V&3PH zX{1k7- zN%oUiU2KWnl4c{+MdlLA#v(i3N_?SazlasYD}uCsqqt`Mu5gxJl+p%mpu;#|?UaRTsH}SRp)>OE>Y`Ma(E#;zsN< zo|DJUxbSh4%$-;Yj?HV$2A*CeTS=@sp2|b+ZA?`cxr$hKJh=`zoPJMCH#cum@F3P7 zS(bfeTQ&0}wh@ao#2xLbW~+&{#ZLO1t-}jLS>zgGFOhYzUD8IKww9O+vR&P*_*B-5 zmaHS@912(W-Z~xPC{42U#LSS*_`T3SYUV}k3o@rhvvbwVo0wS`6xsLT_b4^nK+G4@ zCN%Wm2WW$6i4UTcSScKvO-yfIPfE6tSjIj` zyXt*|@7xEZ%tce-J>EU z+e7RTraj&F!DI>&X?uyC#7;)Xzu^yjWZFJrAF-2zd$-^VN3u|2Gx1b*$|=2=#%2^n zh7s$9ter&5jbAgiLp|+NRBX%4w(;k&? zm`r$r9Vb>A+2x|u=TVUc!A=mf#j|99Lvwy7EZIq77I@rz&F1l0BH1Zon=x&3xQofU zL=<_N*abY?GyK{WRgdiqv6gt;<~Thu*~1fQXNldwwCjHHo|I-3ESlIxoJ+p0T(L#X zjKnOl$RS&Gd^eS8=ZIP1JbB_;YkrX?*?D45ktNT!HQ7^`pEs9WAl3)j$dAX0(b#l? zT_kp9KTLR&JscN5(^1|mTcIPuvE=%5xascv2m9y zHM>nL6xpmEeQbysMN94wtB%L5V(%T7)$A^@UU=NX?hn7DX7`A-!pFYi%TJtEv-`vn zc0eZ!<}O=LOt&C!Qg}e@>o#ERzSoay!IE^=~Bzr>4DF|5o_xfHmHoa)cQ(|%cz~09=|E6Z~!~&5mzMjo{K{73YSPd-l zTlJp2^Ca0bVj;*nTn;U#E;5nWqd*wjkH6HX23RJ7QJ+fW_)-@uwRyEsfYoY)SpD^?6|^*?VH`H$qxRs{BRP9OWwcu?nIrG>?5&hSY&eb-@mBYCt_{^FgD-F3Xy7-L2M6R zZ{t^3T_8m&A&8Or)3cf#|!7bq$2z)5?SOIV%?B! z=pQvwot90^5f69tn{~X2Y7}G3A$ATsIm00Z^S+$+Z)h} z_qindPE3y{SN!a5Chry(9Lo5neBYCCsM!W$Dab4i)<{<~A7YNke6MAC6Vr>9 z_!9ewENMpl3^ns3HXI+ID^)k}TO*makys~uT67p(o?pmFwuzW6J_&cK9>e!;$^3~; z!?DfW)BGxp%^+G5K&%F`BI6=Ys97MfBy7pofJwL2EQr`d+<9_XcZtsuS>$G7cDRKU zdBne`IxU#kP256iS1yN_p+?b?EyP}8TH|}qOg0Au+e&OBzO?A#W>`o?%Cv37hG0wH zoz9u2X4{EH;10&%`a55%*$!e>*pm0Pi}U?Wwa*-30Wp30vq`0zbIvJhfZkR9xD z=o1ww*)C#UcygK9#hNTIM3K9RT}KxB{IkhQL9ji<+G9&nSH3$#Md}6HODq=Cj!)Wb zvSTRNK4MR>CFPbD-~%AjLWvc?BIp0=@>pGD7_s%3Rxl)cDlvm7az8N#99to?0sN3D z*#TlH$fnL|z@KnSc92*tyyJxBM)Pr#>=3aA$Xs4_k5!NDFtM_@J!Tbux&kqyXvq;` z=Ww1(8Emm%&5jZ)fS0NLoU*)MBGbZ&mBS*(d#s+MPKzM+32z;<{@l-#xlD^BHW@p) z=eHn!64ouvn-rpm6-V~`MqCFfQnF*j_9H7{H-Cwm9Vh07tk}alW7O;fF-z=Za!`cH z=71R6Nn*Bmxa+(*%r`2%V5f-v!o%ID{MOb~q-3XwrQy{bG^!$hC?eSzVny&q6}qc6 z@4ZTPmY6FRdE?n#lb%=ZJm7+hT>lJ<(g9Rt?2%XZ$@)EV-w?yJN~p9D6&YTtd7CfPM&J1}kB&**>DEQXi` zj&0-cnS64|v{+&%v6CHJFVm~jt`iHyv=fzg@={GNT5^L}4;))H@0TVU`+~(0>vsw| zS={HzIx148-6WQK0@#y3e)2L@vRlLo;&H1uA-s+{?KUybGmv&*{)mml3}S3|h*d&% z-)<03uaey*raud5SJuY0Rj1t})(AV9U1|(J_Q|yS#O9xdw1j_M?@^jjjO_uj_1MWn zDNDwy*+XI%@wi>co;X&`9uf1yldECU*bi#nd3vZU@rDjivZN#xH9pf^Zm~L6# zr0|s3g5%K1f3kY>=C@?=#2k?A-s^6yPD>!R1X=XUYDsGLj94(Txy?onQnN&2ad-gA zT9#``OfOpUoR|l)h#SR+t637U#(3P;Hw&w+X34}pBb&Qr8ZSd-ktxJ>;Dr-$`#A3@ z7{u6K5PN}VN$F?BO!n{udr9mGUT;&)JbA-ProAE-ibY;8`a3^SNtR0NIZi>3d+DR8 zlajqAwhft=f7(`JMlrTG#8%+g4%b^&PR-sD>xpT-hTU$YX77l-!a_T+_{eIn+DtmocMCbxW1WCpQ7yt>DR&Av%#dci&uyMj|ta!ecEosle)SOBg| zMmn|~rB2Ht_6u3+(MPsw_J!D4{8(^K+UYIC3}S5A#Gc?<+0g$FKDi{zA@&eg<|FDR zWU14>64QLXH~QM`o@(}um=V)jM~)6wv+u-4;Mg8N8O4hyqiD$wVhfP%*wKksc#{1j zHVoO<4i;flq-43oe#5b)jEa7yX1|EN!nB7r{kbJN*SxFSP+0rV#*M{o%|)xLFQYWc zmJzc@R=4rU=W4c`STa8L{r06wf||J!8;K8@cRc&{P_q@p>~VX{ecPB8#Pp&iZp6mn zKG)p?4f*~inLDwYSY%pIM2GnHRCic$wx{o-hR#6#Ta^7ikdeAh8~}E}6Tsb^|I>vO~lc zB6}C&V)7#fk#?Ba8Ei?T=Y#lEmT5=-Vdx`ox@x3sdzNMgx&xYKSuexOc^B4&+6HrZs!o2WYXyh-60v6;xS zn;qnrgpwU6_7pp5msQoCij?dGF&mulQnK8G)a)cNBNq9}d>nUDFUEF?nCA0_b(xN9 zC{41{#D?O9)2d7F>1uX{m6@0v6=gJ9=~X}+*p7@?b`X6K2ywGg=nXk;dx~~z_e5^d=5+<9`+ zPvqTDS>z33_wXhhYUy@?()40%al|TMTCtl|udCTjVq@_^tY6^`d^eS8w}|b;dSd#`cof7My}|kC^-> zQ?ggYmLqFB;W{sRB}*k%2+zr8&5qupu}SutSPvZAt}F+BO)l9RVww+ps@iS6)D+AVkPlS=6v%YlLdw-@)I%5JK-X!yLL|4I8zYL-Q;EdBw|p5LDH*kqAki0#F+ zJ-=u5p)`XSTQ;$wxQSYNTa_AWmP4!@z7t+Ky73M*`%3I@EYg4NmDOtYjhGi6w+7oI zOcwNFY~P9fgY0Xyk-TGQ6zm7FWZdZ;=C@!T6)D+IV#V+*d9fjIxtiq?YlcNmjk&i# z&3+O4h{r9ldOLm&(0S%v-3Bx5`!|0aTS!!YUg1f$jM#EaOPFpl;YqffST=5cTg@L9 zLt~T7l^LehS)b^tW-Evd!*g==@m;(*pcgH1Bj$nUWTsF@eBo|hoa;n)?Ed$%akn^=p>z+Od1T~?=UAhs0KzAh=lUk>P2=S>Pe N#Nv@9<+iH-e*lHph2;PM diff --git a/main/train/catboost_info/test1/events.out.tfevents b/main/train/catboost_info/test1/events.out.tfevents deleted file mode 100644 index 2f443b1b1723b3fcc707674f59bd9c07e8dd4e92..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 12560 zcmZ|Vd00ltN?gDErO~#h8SQv1dz`$zhD0 zjL8}j%9hC3vh}{s{oFnGb$4EW`Qvkaujkyqvpnb5&sKqdXHTrIZmlTlqgpR+c{-IX zR=Su)H_!GJeG+5bTYvjs|HBE2ov^p=lymu?)eKvvkknQ832jrf<|w9h(EpKo{&t~V zZNs*iYYVBT#R;9+tjPqXm5gtzE@LTm;Ci7a+^qkFX%%!~sYk96x?V(`M&`PR)FoF4 z{V~clooNl)MWv2eAatSb>2c=TO6ueYp{?@P9%ovIwzbs$vxT;(V;RD@P@8EL+9jl3G*Rf~N2k|kS_55D>d`?$=QQcd=Ph(8saK2< z`rbA5I&16DE-iJlQ9@50IgGDc=rU5@_ZRx<>g*M)tvog+SXSyrLxiqw)!u?>6?8eN zHx3Z`b@Q9|&2@RHm-Q98e&Ce(Ol#1tAa%c9LhlG_Yj3VAO1;HT=&|m@Q<>JGT}f)c z?n0**Rexk!d1B0@veaX{2;It|A8!k7BlWaSLJ#k8FpsrWXjhTCUPqzhQm%DiS_55G z>Iq~`7i%yl)XcX$eYe^yixrgdoBO5LZO(0>IqPBz!or9PkvJt_HRcczu6#sux8 ze&H^3$KX!4%ykW^6WR#vllRh=X%*V`Qjcma^lu~PI5Dk(c941rYCor2G3L6a)KM)( zdw!PHJf?MMJ4)ThP3Xp3yo1ekEvW~fUKtp2ifQGUF~QnWH+2>5;Aus=F|C5GBXwug zk@H&3HrI8fUWWSks%L!N8no+4-2rvv1IK}^Ep&aUFEtbE9-VrGKV5ZbJ4yYAi_nkH zIM!lq<+(AF22v+E3;o0K#3tstq15fMUE@*MT&7iMHH+crgx6}XB)HOBoaJ7q@pcT6i;#spoYK2=}n z;qT7#qZGQC)W=XandtM9wN+?0m)a6_>gVX^OlzQBr9NIytmKnlQh^-@=oV7hHrN9L#lFsiW~IH)S+!%(M>e@1(wo>mHc+ zc#OIBkh(`LF~QPTj(IYzWE&Gyr9M(q=n(6UYs@v-t(U1G^tD%o&oQk+yPa&$w-Y*M zLHW1l+Dqzvs6!)%$1<%!+gs|wxRUM8P7RsXLARH>iM?2f%ktm(^HS&zQtxpPy61s| zrC3{eX-u%A)Siw)-*B_!dp4ncq_#s{Zpw>X)>fh2NorsGbdnkd^f%X?rA|d%D&@&# zrZs4Hky^*|#N|%VA*OZEU8PRMb?bw)0=pT|-J}je9T1Sh@Ak?oV}jkK?u6@J=jJnx ztwd;FsZXM|w29_-WEI+eQa?hS6;{EHwKdQ^q_)LR{(Xz*&CPXBsUPB5HKUQr-*fo?Ech(lVx74-p{9eCle_^JT9Akogq|U+7nuyX%*VvOPx_gtR%|YG2C1ak@}5|&|gMg<+}n6 z+C!yIM;-rVxg%@qp#7!3jrw5ts7!M`OzK&9l$$cUwqaWN*O=gNsrTbi+MftrWUfa@ zeGJe^C=tSMK=6aOW`%zE2yUT}Z9onO%&R_T0 zdp-}$b%4}&aV7~-Er&3zyfG#iDD?x>E_L*pOsk;BNNtN(Lwv^X&gOco)H$d>Ogh5v zCmOWJNu7Y#<<`rK;#phh@lx->w(q-lb(z+o9VGQuocG*>s18gkZ;hD*OYMU*DO}?d zf6fa%L27qAzn^5^<)4lU?GUM-p`ID}iGN=ubg0x*aV2q;i}3rm2JMMbkHxmXSKDlM zltNFEx<6{aj>!X<)}cLF>XN9dZ6D8{+{!y+f?-luswD0w3%@Ry#o9vuDD}RILMPnH z*ub<3?I}_}z;=)Ho_);qRH+Z49vvLW-=Z~WPm}s2uB1h3z(m&8K~I-DqP$qQ%fV#6 z5}{{E9gXds$tMP}w({PX;7qA^p?1mMSYUqwdY05W9_2+3?FwtFpu?q}gKf*~`ZsEb zZ?t=K@wWN)jW%eFex;2c{neHM_ssch!iQBBQz`Oj`x(Y{s7DZ#62ElJG5f zMRaUy=gGM8!I3^H#F@R##P{R3BQkr8uPUFJ99pd@Mx^>UH9P&;~LcG6TTAb znTalA&G`btt6@FTX8C)@b*L{Sd;s#a&gmx^S8|QXEFyd|-ayxG+p*o8FDAS<-knSI zTX&3c73xa}KZ-nO`}mEFYrvNh{(b&_ufalo5(vJG@Ta)pt7msJfYo)VFDE>IH+kPT zakn{NLHJo*%;iN(mou(>GzPhn@FmFIs@Q*ETm`<0@X@&TnD{vUF$w+?;Vr9+tLgN) zs{91jp#C%AvG}PvdO!R>d^O?A@wr$uvr-6KyAJg=gx^O#U`TJyl~2ZGq6kk%zQk>p zHLD9AP56lX6F4JjDdQ^C*AkwZf8(g~`K&o#NBCO2MK-@?d4zEd>c0?P1M7WiRawip z4tzb~NAZ?A$-R7WbN(ygHprC=uAz)8pN&Dr5I&-UI0=p>j!s})1-^lBE36Odl;m#C zV+mK0=iVLi2jd#lHxhma7qg?QLsQ0e;F}1~e~>wZI@L4hn+YG3Klj{AT^U!t7?X)3 zd=K)~DK@9g`4+-`@FZw=+^QktD%5`?d@dfUUMHam;~Ma-gx@bGJ|?$3zbte9JK<|_ z`>;{jKZ|i4>e~p9!}^e)LsHE7cES^p?_8W)opI%>F~}W+*TF-zI(0OSaTWMZ!Ve;k zFLSfKIp0NiX#OddaP$e|8q{|a-V{GouOp*)UGP1Gr(oS{_*UG97;719chWt%*rATvrjPUP}d!~|;7+gE4Tm|5 zt5Cm8cx|jlwwzwmoF@~$8P^_pd(&gaHK<=9{0Q;^r>pU+NeBKn;cIX)GyJj=SzYj} zgik>3R%rEc#uZCrkpB=~3^(Ks?yhc(tH4tT-+}z6X2bc5kl@z{zk+k$VO?NPph5jQ z;m-J0cKhDC0{a=@Hwd@IcaQ2*w#T!z>%eak-XFPtMgIaDL*Tav|B8p2wd(JmSzRe? z4DvSNO>yn9&cix0t^!Xbd^GaF4IMReeuwZo$mj2F?8UeS^}B=@#h19igKMTRt^>bE z_y~MOygn;qqB&0^JO_E^z16)LSBe;uxled;5BD^vVGOf_~>5MB@#vor1?v4CWLhrZcJe%+d$b-Aa z@Do^t`b)xX@M@a9vkHHS(}2GsybSUa4^DWn#R#55cn~h8P1+2;b{*=k2_KHUtYf(< ztSZeOw!zxV|IK)5yXT^)vAVRaqqxrF=TZ$iRqIh|u%DP~OOBjGl< z_JpMGKAH1RglAygBJ6QF##N|)Cj15R)iYz$&G{F?``}`V)ZLfOxCZsFgh%6#d=9m^ z&7ayj@I1nQ!CyX|Kkx0t>Vj+JN8&B8-f?;4LB^Hh#vo@Cz76YZ;%@Vep$a^L@I2&; zBcAh@IKd+cw?^K*_WA;Elu(~T_-y2v)z$)(&c%jJ~Y zNOm$)GPlvxaoJp!`!ck|@_E>q)?d!{{e3^@eV+IEJ@5N_f4|m>Uoa{Q54G=$H$jzE z-|buDQH_;)A06g69tKrV-M$V(@%@dIdhYwgH2|K6^0RZh^vC_sgY7PE4)|4+cSzSy zkIUSQfJ0ur@kVw^C4NrwX$u#?rD;ySBj8q|rbU5+@%G4|+cIht-U-=-sBzwS9`{{! zCq+LRPC%|hOM<%K_ZT->CQKd!%bbc3&6JV26@4Aq+0YN)Ma}_pmZ3R3yNyyV*J=!7 z;UHAu;2bjq40#0SKPp#j&{iz{_s5=QPca#jY8Su zVlJ=!$}R4&KjR50nYVOs5&D)Nd#@)RM4guLNoBJ4)$pX@i=N22h@z%LNKCmrL+SNS zW^DySY8{u<6Ca*UfMq?qH1A)!;YUzT*8Nw#@RjtOS2E|#pL553qs{Lsy%R*D<8#^i9aKFTk1N$y4}?AFC;ronYB{qD$gw*AjWHStS5sv zyCS=f2km0scdt)oMOI=XMNMgnGJpw%VD2fz67DI42hS-)2z#M7 z6(STNOepsJ31vUo$q5L^*$XGBIBv#&sfQ=UKW!jOx#2 Q#G7P=xT`Z>ise8vp2oE?*!n3)$- TQUoIP^z`-gzyh07S<3kVBp(-B delta 54 zcmaE>en*{WIWI340}#wh_?^Cy$Bso>)72^_pfaE+KPxr4B*rhbJh&u3IXfmfKP7c@ I8A~}o0No-IK>z>%