From bd4310f0dc04c0ee8a8aa5cf3ce45f018997df0a Mon Sep 17 00:00:00 2001
From: whitekirin <113206109@gms.tcu.edu.tw>
Date: Sun, 15 Dec 2024 03:55:05 +0800
Subject: [PATCH] Modification to pytorch not Finish

---
 .../__pycache__/LoadData.cpython-311.pyc      | Bin 0 -> 2721 bytes
 .../Load_Indepentend.cpython-311.pyc          | Bin 0 -> 5785 bytes
 .../__pycache__/Loading_Tools.cpython-311.pyc | Bin 0 -> 5637 bytes
 .../file_processing.cpython-311.pyc           | Bin 3060 -> 3859 bytes
 Model_Loss/Loss.py                            |  15 ++
 .../ReadAndProcess.cpython-311.pyc            | Bin 9711 -> 3420 bytes
 .../__pycache__/__init__.cpython-311.pyc      | Bin 158 -> 181 bytes
 Training_Tools/Tools.py                       |  27 ++-
 .../ValidationTheEnterData.cpython-311.pyc    | Bin 1081 -> 1104 bytes
 .../__pycache__/__init__.cpython-311.pyc      | Bin 152 -> 175 bytes
 .../all_model_tools.cpython-311.pyc           | Bin 4519 -> 3292 bytes
 all_models_tools/all_model_tools.py           |  64 +++++--
 .../__pycache__/Grad_cam.cpython-311.pyc      | Bin 0 -> 5455 bytes
 .../__pycache__/__init__.cpython-311.pyc      | Bin 146 -> 169 bytes
 draw_tools/__pycache__/draw.cpython-311.pyc   | Bin 4041 -> 5585 bytes
 draw_tools/draw.py                            |  10 +-
 experiments/Model_All_Step.py                 | 131 ++++++++++++++
 .../__pycache__/__init__.cpython-311.pyc      | Bin 147 -> 170 bytes
 .../__pycache__/experiment.cpython-311.pyc    | Bin 16356 -> 9086 bytes
 experiments/experiment.py                     | 169 +++++++-----------
 main.py                                       |  14 +-
 merge_class/__pycache__/merge.cpython-311.pyc | Bin 4750 -> 4904 bytes
 .../__pycache__/__init__.cpython-311.pyc      | Bin 418 -> 441 bytes
 .../__pycache__/processing.cpython-311.pyc    | Bin 2422 -> 1784 bytes
 24 files changed, 284 insertions(+), 146 deletions(-)
 create mode 100644 Load_process/__pycache__/LoadData.cpython-311.pyc
 create mode 100644 Load_process/__pycache__/Load_Indepentend.cpython-311.pyc
 create mode 100644 Load_process/__pycache__/Loading_Tools.cpython-311.pyc
 create mode 100644 Model_Loss/Loss.py
 create mode 100644 draw_tools/__pycache__/Grad_cam.cpython-311.pyc
 create mode 100644 experiments/Model_All_Step.py

diff --git a/Load_process/__pycache__/LoadData.cpython-311.pyc b/Load_process/__pycache__/LoadData.cpython-311.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..56b80ce1378091db080aa61562c5adab5100c242
GIT binary patch
literal 2721
zcma(TZ)h7;{9STMUc02ty3H=p){Lezf!fLbd{`~fxnZ4jW0<ZT$Lw*-U9)VtTzz*1
zH&U!HT!ej?usH=k1k~!R$UqUl2!8YHT{uZN7!1Z{{T8W$_{raUNiIp-$==d;zu$Ym
zfA9BqUxh-w2-d~1bT;fq=pjW~%hv=p-UQ$dl99~ik;(}iM`2Iiqk09e$_qThy?LMN
z7yKOZpeZEt*O2T}xK%GgYw*`tLO}MXQ8MrcMU$KpJVUgMqU+-8xxA9}I6J2`Nfu8?
zh9sUL%A!ORr@wtPqiJ~^xO}QYvPz~YlYKl*WaA|Ou7j-#5;z$No)TK%WbPUkyt3yS
z5_s9G_!c~19WVGx=v__-fIU7ZJgv#OLbjp$Elo2re1~`{oxWj%z;>IvgNj@ci7r9W
z0~psri(bH5_8=qBI!9;~z%>^wcWQ7?yJ@Ij)95aD8cgh+mM$oHaXP0PPVWqnas`nZ
z>hw)11%*h4MjE(JUs_a%gGDh{$Qhytth}z|7aTtmpgSSPOkYd+NeFa1J))S&OArWA
zr$LXwTKW5BfKE)D*9}dTG8e^+RLCggJcKMuBsKP8Z2gPt>!1I8o&{*J5lDv6_>C`K
zA^pIAHL}^W<1ND=+mL=3wtrSoEgCEFH~MSQ7uaWbti;>+*+)3^Vc<rfiuYFVUJFNU
z95r#YhJz)31NczJbvhu_oh(Q{0&oWz9e3SMU54P5J8przxr*qL8@N@VSaYeCA;eGQ
zanpMfkGIHeA7n@v?<#0obG0q=MII87TN(pX)<3=d`^^v5zguYq>fy@!Csiq{#N1Fn
z{^r*D_tzhP{Atohs2`o(S;Y_;ktoyhArM-f1mUU?-~m=MqMXYZqM@-bu!bqC?-VU3
zr>`x62GALrYDsjpHWjez2)&-s3@P7n$}-B$pcvUr<p-O&A*ztGSBJLbL+5oLFz7R|
zt)NHYJs$;b1xmgLk)dkjU?p<!-r*l!u_DLq$njENEiqP29IGUb{SvodJ8LCo?Zm8!
zM<2wK)%av3KKawUdFqlCzih`Zn|R-Y{)CAWWDhmRPZVQ`6f&;w;gA+7%bf~D{C{)D
zW4O`YT+qwjqNm&`6@)ex<dw_Sog*9qo^2aJ78yHQ{0Kc+LU$f9c{CxmEYQ=@irUxR
zwE<MPeb<J%sh-j@cyqh;gJs@wPg`Oef@bO|vmw+zYe3yvpY8?S1E*0}gi25DAk@)>
z_BW?54cXgnsD!%O3DZNG+~RSqU?>H{@#d9+v&%g`0o9Y!3KF?OyKbidCrt4!-cS}!
z<W?;@A(wzkY1JO4EUb+OS_3ag`zh_tblf$nGDnM_$=Xy~WDxAWI=ZE#sOrtlAOfWN
zuduD4`vW7^z^FYiS_<4B+;0sg?7>7SSofoza1{?%@bD+bXYX0~fQ=8B_&^Q!S8=R@
zW95M_(_ioY_IYdcs6Ber!V@;0F!2PtAF1Gx@>DfGQHf7j@ktvhFk4N{R#LN8>YSZA
z_nYvxIe)>PS1qAn3k3^nHr7n6fzs&yYUEHQa_HV0lqFqBrLELiJ9XBao3GBPl{wX#
z)9g9TiY(fZ#Zn(|gojN$#Hu-o>7Xk}sv?R`uPCaTyp*SKP!!);lJbokHi^g(jWJI<
zAak1&YR<fjlF!4K<j5}SA=>g^WsQr_N?;pJ8T|rmFsWMq>K?x@R7cxv;T~VIj@<2;
zkZ<@NUq|qHijBRBDLoBud#-b~jxWG}lEh}Wgaje9r#R6rlFr=GJMw*Jqy4iPKh9t_
zHkhKjsG*cL>wocN!-Xkw1W@V>eF?U@m*cn^I%0PJYUrf7-LHm5%<f;E-^aoGw{AC!
Nf6~*xwz*<v{Rd*Dj?4f6

literal 0
HcmV?d00001

diff --git a/Load_process/__pycache__/Load_Indepentend.cpython-311.pyc b/Load_process/__pycache__/Load_Indepentend.cpython-311.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..4999f35651739f452a63fffe1e420cc96db547e5
GIT binary patch
literal 5785
zcmcIoZ*1Gf6({x2v@A>ho5oRMF;%ogjjcqnT_>&6Ot6|XaUwT%u(nECRyAowuKu_b
zU7&^p1ZcM?K(Kn7v7|#fs9T!3NCUWAR%BQM<fA{50s#pG0vOnwCf__{{jyJcca$j7
zmWvj{&eZwt`0l-Tcklh)JN~J))lNXU*f$v;auUQpP^VTb)yC7ep>d6136@L}DKbiu
zsBKD`Qs$@`x6MgQ${Mxeww}h4w56J&O(`l$k%WmjL9o`32-bGXL=d;(r<Q0l+cZh|
zsVAuGCzYKOT#RKBsaTw2VriCnmCsCZ*{nJO?;Y8TbJNpFj$va$4Eoj-$HzIpNohZw
zfr()lg*n`8jOP?redt^!lgv(4W#WHiwez$O8rKMph>|Q3HL>JJMAXchpti7PsI9C8
zY8%o$+QiyG*CwSKGd`JSxmhkPaG)lVJ7w1KM+-##r>|i4LQNd>Zo%(1{M3>om|By_
zk<im9b0+9%8^w|!+BAnI+v#dqyZ_bNhaayMF0Wnr^S4*N_~+lh)?OuJ?{LXSH?ACd
z^4ahE9^LqGZTa%r&%b_r|H`AnLMF}iU(5*oTzV?QCerb5fA`Va{XeXI^RYG?@jrjC
zj}E>`AEhI&jxuR39upGpb4H3_AAPV!F`LOI3=|Q+nYV(ritRMgC9AZJr@2=$0`oG6
z<&j*!S+Qif<TT#`WV{2aEZQ=V{oN@155I6BD`Zl!sf)~1EIq~X7r+?ed@M!x(~oan
ze*DM#7w|dEswSd#bcYg{{Xn4@CXr4E3^U)`@XP|W_Y{)uh3XRVos-z+{?%)XuN5p;
z?Uiji;a^LIK3K5G)UFEUC|I5X8DThWG#j7xTX1^Uh#XNwB%4Z(yan%XYbSgGPS<ea
zy7LyA>eE?LXs-7Jv|ZJbBW5%UG4eL{=7F9yq1rf3Zmp}yNLkg@R+r3Z8X9ype~+%(
z3?px259?}aq$^8+4_G5}E>OZQa9M%DW}XE*pF>FY?VVpd{>$Y@zx~8M*A3nGW61=-
zL?V-})6sKi_vkvMpHjNxoWN8KhXsdL!A7wl35rgRVx8p^X+be(1zs^HxwK-hi%{C?
zZB-oDe=EpRoad0H2(K7ElM$4zpVjr!Mr)SO0}d04qiWMkDwaskKik-J^)y{DEBgsl
zmxzkbU-lg<`HnqsO1^Q~H!e}nf(bV-3@#pfNVSSi`tyO$vrB_t9+URIDDQny@{Y*f
z5y?4nFE2SyOVk;eIwMkNR)a(3;8-a*CZ2v<3dZDMOriq7$926+?g?N<zB}eSqstcg
z;LB3qak=lfwC{wx?}X$$u@aD+uS?V$GWCX7d-yicZb~y&=pd8EPUD378nHD*=1p-T
zZw9B*90z7<WusH&EkLQ4Q>U~T4Gk!Pb_VHy3rqIYnnb;|fyPvSQ*Xshd25^iEt~Q-
zmdsN*YnXWRM!q>`D%Rqt#u+sXF5GO?3N)%a8C92IjGK9zfV`ku^sC6}=CEeuMnZdy
zZi8f-QL?eOwHxNlMI9W~dFM8)1i+<f3(5_Wfsz7v0(Z$%S3iZ-;JgHJF6zKeC&|o`
zto0V0+->dj^7bDvuc^Vj7C1KtoD>Q`n@Y}Z+!aiT+<H^VIdUz=GeVG=cjRrr8+rAn
zl5=e0?a0}tO#pJX2#-wo39|S(6h&#kxQ*CFKS4Bp7mxNp={d0h#W1!^AjH`|Im=BY
zrV|{?gi@I~K)CT~oU-7^tGH6JcR3Zam}COL?HS;Q<1vq5l;4Kf+hVhj?67J`cj2r{
z#WVy_$}?dEZ=tz(3h+~f+><DWQ@Yhu4(n0{EQtI{YdXWH00!qVt-n=8%C3Z(QehBA
z9I^IoDhu6g=TtJ5%_gQls3LXOxr`7?GU3FOpk{!0UUl>|tf)T9BcxQEqcsLNH&Mt$
z73YP7kmQt(44;T6(!df28I0PU#xWdt_)cJ!Mb-d0OTAzW=R06{et%==ZWwUfh}kdU
z#4iyK-NTami0nQhQrlO1+y!UF<1KrRmOMwrmnJ07r0kg#shxOmq~sY9kDir06S8Ll
z4bp?l=kA_=(0=zVnGTD`qvCl+K0htSFUsc=^6?qb^Da=W`t~k$EJO<SilenaR(1r|
z30rd~t_r6rJ^pgfky6i*``&vqe>?Eaft5jd>}@gjjyyIc^{{deTR8R5(Y~-}asTz$
zVxTxIbv$=x@~@rC(XTvrJyL&2?hi>Fp$BKAj<b?uLUv4uj)_&rj<RES$+5fmoaE@2
z9sQ!CztV85b)o&r@j|E&THUq#Mzp+ZxU_5do^yp<2}--p$h#mo?WpuVdt;{Dd$`nl
z_+IZy`$|aajmW(bk#euP+6$J7%Tsm@mt4c*;Zu@pOm>aIn0tTO9V)p)5B7+!oR{1e
zWcLM$+Fq~}W-xhw$<@EaO0EIfHBhjux;hu4qRS&vp3Q*^!83BXr>EeszCUmY8a!>x
z`yrs28i%Pkq0grP)U*&{gbO3s0EC)~2q`udJqMuCf{UK;7IbcF11xETR?J(=-83?2
z80SGiH5XAXys5y=28G)CN?>1f&~43G(d{O}dtj!&nMXSENTAX-38;p_Wi*4XT2*jD
zQ9<d{-XYJ2G%v<~ptK<v)K5@awGl+7v?Axjc!d1gZ>uJ0NpyYGhLxt;C9)ZNFuFe~
z-3CTXRUxIViarPd5%6TQNPQI!R4AaKKYySx6d<1Z+!L#S*c^sQ+223~aM0nEI(%}6
zuh8_c%PV#H<St*Kd6jaNDY`__#g5M>m%6@mOTHo5HzZNd%hdBC^?U`r_m(JcF?@4u
z`I&O?WGQ%33XaLav6ZtDH7--*A~jy2_m=6Q5<Mi*!!kWA!X3xGr|dpdavxfLNgRD$
zjJ_pDImtaOyQdKvZQ&E>u&_@gAto7p0^Mq7wA5{*ZRJS_za~Lvxf|#kCR|2O_@3G3
zdID8%8E3zXW%DV$*ZprCjr<`M3}A29tRMDDD?enlIo&FZyh-Bf0r*Z}++p30jewkV
z@&~R=H*c%GGD`nZE7Pr^&izNOOveMQEQh{{SX9)&`4Zm?qkb!oc3S<afKnO7%E!`i
zj`yOM7S-P^2qF}-kecNKFq}m*kjttLyBm7*fyNHIRW3D(*tDzR+vUC<F7COxf9XBR
z6_i~;2tUrYg85;Kvmh*l7sra^VnlX%rItN%%bq*trSNjsUH84ScYEc&pGx~e^1e{1
zB_y_l9(M2ebl|#x9hkfskv#*Fdr)={O5H<p_s}{)whrQIfvmK3UbFw&etocLx!JN5
zy6KeN2bM|MeMo8>klO}C{8YR@DTEig7Twp+F7`?eT6EBB|6d6eS7)IGN6-jdwBc(^
zI+o%XMzJ$YD#OmfcRWUMFwA>%v84KjN4FY&C#p~a#nY%l$MFS$*{J7(t{*(A5bR{%
zf$Ea>w{EgqTh|G#+HSQ*$#p_kJ9k*=bwaCNXhvPF+SX$2S|>JFs(2(c;<qb4{GAeg
zp{!lQah4iD!=W_$s`^bz&CF?7DwkA)OFfGX)SK8|m3B3$#<^_(x7B1@{n}Wi#zu*3
zw;m($M(|-!tiD74p+?mGeThE_FA(}>KZ9!BOp;`Uh=|6gLL3o|PlfP`#;4J&ut+p`
W)-5MU685lh{VzQJ?=9ZqCHxnVTQZOU

literal 0
HcmV?d00001

diff --git a/Load_process/__pycache__/Loading_Tools.cpython-311.pyc b/Load_process/__pycache__/Loading_Tools.cpython-311.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..2cbc0d0883ad4b9c5a0545acc2d30e4ad7053474
GIT binary patch
literal 5637
zcmbtYU2Gf25#BrAQKWpNL{YY+SU;yf+w}iW?Y2S?+l^Ait|B;*5CnpYE5kd<R4G!~
zqimR<3Kn&0rEwv-5F)qmgMg-r1f#8sqNq{Sdg)8hr=vl@!~q5ZG<1|V22N0*Po3E#
zPdtit8u!Ti_HTA~c6MfV_CE6Y>IpoT4vtR#8*=}`LirI&nJ4EVvrJT?axpS)rQA5j
zlOdw=*NEy`<q5e1ul0=ctVMtpv8u(Ta-$^V{u~KMA|-m3cuIvQ*C4Y@G&0VqWSm!9
zRo69P+@->3<AN#zimDsXt=4IEss~=L=9zG-^+4rScR*Q!*9Wg3>OE=$)Yq$xTHOS%
z1|YvfZGzsC8U*yI&47M&Ct!oR3$Rgb0Su@-aM@-Ch7(~`84M@G$_1*;gsEn_sg|6j
z@sS8u!91?q1@Dt_#``2m6UA<3C7U{mZMLLX345}KQI=&+SV=6msbrL_b0Z<Q*{N*f
zN>PTxZ)-7SII1Vjj#}+(4Xk-jt>%{#@uU_>DudBTGMb2o=~c707Tx!w@o<b$DJn?C
z?5{-xq8bff)-1Wbkw_#>xfU5?iA@|&u5+fKYq1F(<s^S~25@lTqMl4lha;DiNH`wR
z=*0^(F-gPI@*#Qi!|R*By?>F(VTL9mny#}#G(M?}B@!|H@XS@yqbSjMG^r@57q$uP
zaJdcTtA86HL-OsN^Fq$kmfzihX=}c352oGu?p{ng^S%9;%K1(i)7^QGG%q}Xt|Vgb
zAQW)_m@Z5}W*K&I&Xu%xpO7@a3Z*-C%^>+<%bbvQt=Kit&3@9HlbRzpTGOsd4MM%3
zb<iQ?GDRAWPlAeuXJ)jxN_|)>(gsM(x{{zx_plb9OkOfw$>|xBi|Qx<86Vn+bd94L
zXaOpGsY9w~yYQ-M{LoWB4FE#)H_p2r1$Hf7zS*(VQ6!vKW-xy?FEuQj|Hb*mD_N;4
zCv_R@|4C;zlcnL=4S(7MV?=D@fP!sU%}`h-X|B?+?4_@B6un1?x7J#}WehqFP+prK
z`QB<v$+cIuMX?h)>nJ`{4yFukY}5Y_KD`xudTpb0u<5hKi}sr8s?DX{x2=N7-F+y{
z)-5JYk*$XA0wS|<ERh83wbnNpO=%I-)$(KO?N6X)nK*RT<<vsZ*@K|7Tw1`%ttZV_
zmZIF8m?l%UPOnsMYPkzZi_URDyAQGRfuDT=hzfVFZ~poE;}3paSe*a#{$C!?FCHso
z-bWWT5^_=02#QCS>7JcY!$~dVX4_ADu>luM`w);tQ%GvlGrAkHsOn{f2-=d~@_VWN
zXUKbttOJJ9e*lmng}^SOWp6gHFBjNXB)s<o15o(pwp-mxuim<{^3Ls)5!^pNn3o!J
zQvXBg-~;L4+E`XPo|BFn>|e2C!bo^pGfO(myl1eak?7)IiANGDxH;2Xs?_2U`-}i~
zFweY?A7o#(i-J?yu!|>Qm$<aJwTmj3f&pMJ9kpBSu&?g_VPD<OnP6W}IrM2eCr7q%
z!ZS`>mE1k|>_=4>oae&qKVWh7mIlZ&%2^(eH~)|+EPPP7{<lvze*W2SuAyDYD0avf
z7Jpgz{j%M`k~p>v$dBL4Jih&2;r^crKU=Vr6zDhq`p3e?%_@4<tGsOkvZGkoxPjf6
zl2>=r;w55a<in2&cNc65v%cbN4O)4}xVC-lfCbY%0HHd|I6_#lAK?JPK?K}hx)(sl
zbEG65>o{N*QgP^`be?7*Rb*&<`YJ$%eB~!iEsIyPf$m(O8?ECg12mD=TY;saTd{2O
z{<ZOY!oAo3UY9#`#t5D*8B5PYDfB=Jt@UN4BRT1a!Tw-y&lt?;JdtfaVVpYmad+<2
zun~OiSwj`YGBeDlo4#`Fg1f9cjv@h;!YpeLGI$h${j|^VU=2{Vo)z2Qm1{7qKJmdC
zI_#qh&MXONC>}N7)bOyO#{f)$aex!ThFF`i6WvvrLFWIFCbhS=!`eo80>;u$0>Ek3
z4BNnYk25IAr|+S3;DK~tZ7?ex%Sp!!_Ak21RMweSxj@V_fneYO0;v!L7#lbHW2lyG
zodst;L!+0$Pum(_X?Zr?XA2vJ+iRPvKYhHkQF!-vpWn_v9=b67maSsp-FXm&WlDBO
z+79zk7K$KeC7Jn$zV^&_QeZ5oW8{<}#d^}%oNJH`kNgc+&jK<k!zkMr^5JxB^PYCP
zI|yDvcF>Ap(;9-ghAzX?^*qdkICfS(gB?8EY6TlMJ4kT0ik*FiXWw(O!(3DKb_H>^
zk|V(&lrdG~M6N-GW?3nze#6uMoUAd6tY&Q|&biH~YFS@R!l5em_8OkO&&eKY=eCV(
z#95hMJtJ_hron4{rObK^PY+{iou?Q{jD(u-8Wo3=Qc>X4QKl2>Yz*^~qP#O3j#(`<
zfE;1SMmrID08Al<mr++Vo}~M*@+iW$5KbX{2Vnr=MFiAydIn&fFb0^%qXO0<;H&Qe
zyUy6ZB3~!^io^*`qS#X;Hu!lYvB6{KSe-`{eQSM4?z?9t#^{@@SPg!kcyz6~NFcd4
z$P(kNi&i={&C+V<<i+E6ffSNzfN8<|+TiCzydsss&50o_mH>)-uvh{pLX2qx6nn8)
z4VDN|Xc2hr2fZ~uPx=5VKC!q<pIGWtP7~lWiFi$fk1Y5J#A7yWoi6NVZRv&M>xX?1
zV0?{ViApz8iaKAds7zN<tyQDi_wf<v!om>ncX+>MD;aZyxHt<tcmBzyOS|A8Rv>&7
zXipMq%FowcJ|M(Zxm6hHPFXHw&K;M(*%&YaTy3^Kea*R(7(|txDSPB}RY(kBv>U=~
z5taYOkJ2b>eipC47>jN!TOpwpPnz8RkYMr&-4tfRaB~-?64AIROvVy#m!djTpb!X{
zUKL%Y3U1o;M`B@Jk4{`wVt8vX#i>~pz6^-0L~S8EVm;p^cn~%Oj+W}E-Xcfy>%gi$
z36LS5>;(05d%=PL=3mJ-fT6g2M;^7bEnQho|0=yQxYmC+m~A<kYdN_nEDD8+hTd$;
z;atn%Md4AfW$C-O#%`WoIt}qiSMU6}MJ+3}8&Z3|q1EuTS{jYVPly|Xo8I>zM_)!5
zL_h~*3b=IoCYG7zp*Sr+at?C{ufeOo3Q*}sICK(DaC^nJA_4eVWyw>);tPik2^2@C
z7Z3ylHmj49|9D=sT<A;mEg(V5(ea)GaTdqri8sT3d9owJennx3<3PH#@e5Y}Wo7Fd
TypIFMx5F1$_hpR+wt)Wv0}wwE

literal 0
HcmV?d00001

diff --git a/Load_process/__pycache__/file_processing.cpython-311.pyc b/Load_process/__pycache__/file_processing.cpython-311.pyc
index 7871aced8cc2d052afb900b6dd24820d2f196226..e103f73130081157f2a1a3a208b057929aa5f2ea 100644
GIT binary patch
delta 1280
zcma)*|7%lM6u{qo-|}Ail2uJ-CA5jNlfJHBOj+kTEsWqIh$~%E#Eg#c$a`wA$qV<r
zT3eKkPEf~S4R;J-V{|BLH#UDLY%nZr`VTZxSb1!qKcxEGBI`yGfAE~w+A%s|_rWLU
z+dcQ(b5B-xvexj8P$<a3I@Osu`FEfao+hbE<v{QP_kPf905(+a5B2?(gos0LjX17d
zH1el%d83rK=*SSYPg0{8?}|UVcJa~Gg^{DSVdl!z&RdQ%!qVhtqR;Q(canE>VeBBJ
zs9n0Y^7)S|-+Z?6^`(_7w`w!jYu{a|eR=bb+n=lpwJ*M^P0zb@%t8~mFJF8({Zq;q
zb*$kO>k*o^H$obR2gKtP3#VarOz0vtc0q6tM!?`HMdP$(7N|qF#J&Xn$tEZp3(Tbo
zqgMC35S%+ypv~-c@v@8wXcN0D=17$7k*-EtFr$z^)9VEa#c^72tvQjZ7!_p?q(Q9}
zM5hPvA-Bv9$hR8(4ScEWiH>o`qPy7>`7l{#$v|R$BA^nD%W%3p_$1!l-IHcNDDBD|
z@x-ESTI`_`lec5$7N)5!+dR=UT*Ff{kS3S5ZI|_{sh+)f9$h~#^2WlBh2ytB_;q0D
z-oQ|GAiXq@uJ&h^`ZLu?X7O0I8p*Qn)%Tk70>nRGG6<;unIx<dn2YkJFbV28;W{++
zTWqYq&Wmn{8A_|t$DNnXNwWd=ly93Noaul6jzw#R+ruR|M;zzR3R!NP5H8CR_|OCQ
zr5%u!z6O{SyWO)c?L@r`peq#HNB1ETh+f2cBS7**yKI#_344Fk(q+H-ULbFmT&v`I
zDn4Rv&?s7-id`Jd!OPSA{>BsFa`3)!fBEC$h5UXFA0K96dDnjU^~V6qaef29GycD}
z?wk5Ov~8wiHnSMgt08?Uq}Ms3#%~TUYmJ{C`S{4pcvXunX|Y8uRu{nGdwo2{xK+eI
zwiilkYilp?wQuZquvNb#UQR}YI}tJ29JmuDK%J)&I0ki%slkpmzn@_+Q2Y;2oKo6`
zXh*z>c$*o)-DH-16zrXnq+NAx6Ja97po-^^iIhNf12Qk|t#g}Dc<B{XH=u}8XPw)G
zCP{GZt)Lo~AOc_@BCnzn19(zttXQ6)IQIUCcyie&nTF%>Ws_waV&pKJXh`*S5Xia}
MkI}aJ9}fSxXYd?d3jhEB

delta 781
zcmZ{h&ubGw6vuZqyOVC3#A*{uwf0A#>K4tmsp6qR50)U*CJ{Z^!?Nv8+n8h%-fl}Q
z6pQpCUJP^TK@{x43W{JAFCP2@ESORh7KC1EJx1uoo_rHZFY2&-nD2Wt?|tVj@80i-
zzKBFZ60WU-Rd$iqq7O)}RvU@2(A`!M7HdYfTUV4j;o8?+;vi*|9M51zCU?!qL3tLe
zl5wMK*=7aG)68`ZK1n{2mm6Jj2RRK<ysig)x$OE%#q#ES%Bp$SbKy9@sl@qOKque%
zX281|$Af-L$to~A512;j%81@x5cZ%Au^7jOqKBEz7g=%w51V6o=-^ArLn;cuNgh^B
za*7w#HPX%7Xea6712hvpgGtvwEK56lg}!O;#=S2)6<=|!1qKO}e3ZudWY8cx{84a-
z`igCN*2c%6MlR7?k{+BtfroZi-tN9${khKfKD^s||M<t#CyS|M(!g(`Y_p>2ft7di
z&a4T%-f~{kiLaC}k>Nz^8E(e{(CkdTvN`Q8=qhN!h&F@oF&(SKV4$hRouY}8;J3B3
zaSmfHAkcm6`sVOv_4CrV?9^6vs-CsCvvz%)ZI82hjD49a)MEuc86M=nwZ4!pgNnB@
zdW5GU1IcFlV;E=-vuOnSh13zgT(Ny^44<6FKg!?=Uyk&VXZ(32bzf0C8`2TNMD1xv
zN1#L<67mo<pq_0=M+gPfgph{_sp^IGsaM5@gq!~Y!orn;&|4rbp-1Qy7FB1yRJjLY
n955&Xv|>57<-!OGUXAvUQN9{YU+yPZ`)~XbvHgD}n*II&QR&#R

diff --git a/Model_Loss/Loss.py b/Model_Loss/Loss.py
new file mode 100644
index 0000000..3a09fe3
--- /dev/null
+++ b/Model_Loss/Loss.py
@@ -0,0 +1,15 @@
+from torch import nn
+from torch.nn import functional
+
+
+class Entropy_Loss(nn.Module):
+    def __init__(self):
+        super(Entropy_Loss, self).__init__()
+
+    def forward(self, outputs, labels):
+        # 範例: 使用均方誤差作為損失計算
+        # outputs = torch.argmax(outputs, 1)
+        # outputs = outputs.float()
+        labels = labels.float()
+        loss = functional.binary_cross_entropy(outputs, labels)
+        return loss
\ No newline at end of file
diff --git a/Read_and_process_image/__pycache__/ReadAndProcess.cpython-311.pyc b/Read_and_process_image/__pycache__/ReadAndProcess.cpython-311.pyc
index 5efc188a48a64ea3b84d3c44c71d68177cb574ed..07744b3c0fd56e29b495467213c3b5851606481b 100644
GIT binary patch
delta 846
zcmZ{i&rj1(9KhduTeo&(oq!7BmW{xIH5fghF)~Ro1P&yN2IbH|>SJxTY1=KY-2pFL
zNIc^^@CRr#(L@hB%fW+ZFJ2n|0Om!AN{ko%emjXV>Pz~0-{0SFzwM{x4`ceXY&JvS
z+?!sh?`3xM(Tz)>?Gl^V)Fo#BrY2)#iP&tL*omi%kQeaukD0LP3NeyBRI9aL*7!P(
z`8)&aE^&xSZDKN8+j(Bhs@ud=Z3d~CC@<T^yi{pr>gzIuE?ip+!j{Kt4Xeienj_Y3
zidJ24uXLfb|K#ER<Ja;wHL22lMHyv;$u7H2Re8v+Qca#u%+Pe#O1z_6S*T@bLWfJ_
z5GMdpinrU2Z;L^+3?YWm%=~twa3T1nkrpbNPa;OpN892QDyI>nfFKPh^^ed%jz$)E
z$gTNK-E;hqhs~C6$(PD39qxWq-qIl)rS}6)F9DK~5_$P1J$0^tN-}O2C2gn>DK6h|
z+_+tk1TXdwnjBQCbX?w5*Xe-VQ;){;;1UyvNkkcO7I9orzSl~yjU#Q28uD_c0eQEw
zkM2Z@;C|hSV-2O>IORHiJ#0kjTFVb%`Y036V7bj8jA%1Zpe~L9bAS}(JbvG?euq4m
zhObwUi+YYO%2oY_J^)b$6PQ*tH2H0!T){s~oIw;32I3;(3IdlYFe`2#@U>&05CpO@
zX{{^D*a7(){Va@wE#h~M6u;xOH$}XkxX_=)zpE9mWp`X>PRv3$#SW`-t6Z2rNG?$d
PP5#Zp*#6}|-gu0k;5oZh

literal 9711
zcmdT~Yj6`+mTtXeTekd^vB4m0@B?gvO*Rk*!aBT?0Zb+{`xrt>CDd(OAj?X(9Kew&
zhes@fDa^!x0f#XOam>rGvyd@hljYCW?4K=}auuqsauvJdmaJcRD%q;k{Fy!HUcFip
z&P>hz*;f1X?c3*ZAKyLq^l5*SnVHVO5h&ZweNf3Tf54l_p$-?m-w1_^49l<z2eX%6
z6?>I%uXL!|)O*zmM#=18Sk(!JRgWkc<`(?XW3Pr){DP^{gz&cMhC<AJx6RtZZMC|s
zJ2)rDTixxvEzC<U5y|g=2DcX(j@heV8L*WJ+SPkitoj7ASIugGYFI5$Evo~nW3$+_
z6N<fhHXZKL*bJcQP-lR;3|0v(Fp5jw=d7&7j^-^^Cu@0+Z?|!RKx;%T&$&BzXOm44
z*BUIV;O~1aX$;GNB1O{;Afj5}9L)kUnE$kXTf@Gl_qOZ<jV-*jZ67g&CS36*NOhx=
z4co1C9TD{wi`{8=TP&R`7iwH>giQzL1SH*zUvKcLzlS1YiwqgwXyA{u_}`%bQ>N%r
zrZMigJ{^dbJqkBUgpUJJE%FKMQpu!IH#s~ikJ6)TRwi((o#kbWTOZ+Km=z3P)TMD}
z#A-a65op7$Xgd_luK-A0nxh)0o?Z_r4l##SzhVw4JZczmRR<3L+*o(u=0NB$6q~BF
zqRRGhy{NUf@i3{P;oaTuZ)@CY*|KNXp7(Jo1$!qarfoZH<6LfgyHnJ-c)QbGrR5Rq
zqS|G3w~BguTZ@G->ab+xc`<`ZvgAg=324_$4jwxo>RULsg$3;X#C%2A(m+VSQSN45
z<e9wsd8H<;V8&E5a9}w9Yhk)#{gi3Lv}uEPhg4iLv}*XDMh}lK^3}dQRlH@oc#C(Z
zcju2Y#zpgtBCQ}sW(+y~%X{DHd&e91`w@<L74YFz%VaP1>5GK|Fnr?mH<`!jjZ0Ne
zHO9sw)zcyklx2v)xgv;J0yA=vX<@pQ9>rGXLwA=7(4&4OdQ^WAdNdy8a1=9ODSF6I
z6>&6-N8?d@)QPA{#7o~Ez!-5`R2{*X)}0wu&UmzmGJr9rOM6t?*G8`~jJb_5xdTz;
zdX#bvfJ0>ykE~T`d^TM893;6wM6K23;+!m>hc)>~aOg#)(<N#ER@Ng`Is!c2gwKlb
znZm^v;>{u?#X!iirdTsmuoa?{v@~|Kv~f<i6=7^4zO(bS1%1UUq_9Q8mq5UEm=<|6
zUgQ9#bZJF|cpLqN#ol)So|#261d2buqnfVVFjf4<bnzSgx)<4tpJ$go%P#$T-Svji
zmT~>RRr=QNoGRNjUAAi~d-rtqZXY>+oXJ}PNX-PT5i*lq&~H7bAJF&5{laj~EP}%S
z09<qX4i9Eb8COgjSNQZR1RRu8jcJYf%BT71jq5c}ml_-EHBaj`P`1UR3D+mV?(d;M
zyq(-zMSYA%A)n3RsJ8>b=pjRN)Q()TOW{!rNAVgJg#xPtpQvJ0Bk=TAl%q=_Lpu6r
zawOslR~`COz?r%OCyV+!=}F(*rSrl5&!nLX!QLJMz|f!J2x0rcp)G}rsIv+LXMjY(
zeZ<9yx@Jea)xEA>)PYB9wYoTd8M3Rx{tsQ4sBu{La}EMS0h^29qsS|un<R2YwlKT1
zrL%rP+@+K$1HHm8frK64j4`LLb8y9!aoMzS*>hw0Gh_MJ{PoT$WBs(T-mCT-^Uius
zc?RDe){j=+Yn(E^K5cy6tDecu9ncRJoHGs>y&B-@{|E0U;NW;%@G@V9-5_}QPbkFs
zvfT;0SPT~``Rx7;xWKtrbiy{#1v_JEgm@CHDVV;dzYHD3`-J`sy$Vk`zY@xPIg$z_
zB}n4u7`8?TbOccidz&Dtkjf`H)7j3qSsnIH;#xXaEjYO{-qm1BumAy9V$6YA%Y*f<
zN-H76yJaRfe_++%FV3wQSo1u$@>y=>o!r~SQ@O8A=f39M;?FG@RGq7Wf+73t&Qm)F
zRa1r~(}pEJa!LS?!-MMHqkTtx`Vum~A)G54Ha&7g7yW-j%|!->{UBM+d|C@E!Y^Ud
zf~yLy8mE?3x4=EC0jh}~YIk&bGmJ+q2Z8(!w!)2FS{XltB_1uhaeBl`a>*{8he-`6
zj}DeUv^f?h!{XP(>%@q=Z<7aOXKfQ-4|i26G0o<%3WB})NL{R0A1kf{7Gs3PLRNQO
zVnuyo#X9~qP|mNX7Bsxo*}{n`2j>)xZPtSvRbz431$UL6ZemryDrUIb-Bt&EDrydN
zuppH|90su~@T-tZVHdTas>R(Z;Kqs`1`q0Gj5e#o*5R<a!$vx*7ldE}_dB3ZI0FPg
zn33Jzc&fqM;4g0Uk<+`=pTFBjPVYAFwik=b2Wz0Npxl>V;oT<Xn9d7>Z%!3fOcz#6
z<y21RRL(QXv>ggcy^Ro|4XzmcWVCMR=yY-IlxfYhX^nTgzhwFIlDcOlb@$50S5K8}
zpDx+%-P&K!|Ixsr-X?F;3uB%)gSw<RY%z#j;*$Ou3dAMJwy4C73D8#9rS_<Xqb>@>
zqDSJU0S;u5M+F#BbYR#q_toU=<Pe#F;N=sN_u|~O(JCe10FQ`+sM63GY=jC?XKQ!5
zA+iwDi2#{m{_o+5fO}MIo_K`fPs~jMQvMOOya5EfenFu(eJ01$|M9t!0q_utv<gBX
z;xL$=8%mxTN(Nh|4CT{?avwSU#v-4-DA~1RC{4yO8%73Jy&}5mP4VlRRkE-Jv#{df
zr@gHkfw!ew_{KyI++<mO*jvcu(m)0!)8Us|E+Z_LEwg91otd!e(MDF7w?M{g1RH8L
z8?5Bq&K&_ib_;!3tYvvATb9ojgn#L*>2_gKk1%9d)KQYfFzIC0fRH#b`hJaAID0&B
ze!L;dfQ-$Y72bHcLw13)*Z>F|Z*DM4<Hv*ECsOysAG(i8_j=|g#-)>AgJ|IM%c$my
z9(Xi2dM7wJ7CiQ7?aPV2P<MZ@_o4LUL8$w;bJzMnOlCx|wz(V&+X&n}D|Md^jD8gu
zx}CyCqQ>}U3vSjT+nS&UZ@0HQIIFWZbyq@{uFpOWcPKD=De&lm^t-XYsvD_nCX5>G
z5gj{fP&)Z-)LLSPI&D@9=MwCWb|*A{@0Lb$@Y1BzcLByCQVoNSZG1d9ck>L4dmUHv
zdMYi4TJ5l+AyiA4$iU!TsedHU(;FNZjp<L+z=KC8Zy96{02P7r_hIINEp&M_aN{?D
zi>K!%dvS&yjX-_y?g?q=OrnxQcGlf$5nAodZc9|pgR#)#t3=hOv*X>tZ||V^gA)Po
z1!_Wu7y?a<UIC0F#{yS}Hr7GkrQROs1s%-b_@s3AHnalULPiEO&2^uo>~C!f^TR+%
z$Ik}u4^uQJ8L6w;3SGp8U{~=vJAMaL(%woZkArAp_pou0NN6z6KKX2}=Vs`_=b_63
z=E|C+ZdaREC)HJ%XCIyh5wJn~4X~#zOl2lXd{cek(lH#0vv=o?PeP|)K&YM%YU--Z
zYr$mQhwyyP+ha}^j^i;0uZ&CmeJ>}j%ysvrV6CfNYfd(01`V~uNbTDAp2Q|szVcf+
zdrPbP{|^+%3sjB6evtdy;gH;Sn4ILw_+cb#Z9zaJ%l|cqNFn`7NTlHXzrlpi%31kT
ze&LwZdsFiENE5$<W&Lu(3qD$!_y)?M?#}><p`I@S7amIe=P29UH@_xqy@~6lv%S47
zSv`1%4;<}>IKJB40f`#^LlC2jJUBiU7<dx(WYmG+vWkEjmXlo6yvh7QT}{kRwYhF}
z><%@{{EV&D>U46BSFeI_Lqgq`L!W*Q?lo}s3b==8i)6~0?GA^ft)1n-H-AzJ;hW&L
zI^Ct!=29mND<_a5&ow(Zo4eH9&YMdQ@m5!97rGtryV&c<4pI@8sDR4wK6UI=@O&SR
z`*qxT&|wch#9jZvh%|<Ke_-^QG;vnCc1t>W^5w*3S<BEZRzw<9(O^CpK0Ma_Hr5ll
zGgsO^wmLQ)tV))RG^fFA4IfmL4Ijx8;eLWKw87f;Tf-7i!kT0W9E>nKX;zdZv12oF
z|BIj?x>3Ps!4?|0FcG4o_3(6bv|xnjUKBk4tvNPRo6NC^LZ_Mtkk)qI-U)#?yh}AZ
z?5>23JG}AUyDvTb*Wh=L0UwF-A9K9h4zEV9mX@v_3tqXOS{gtDvWTZp&l$QHq$|VH
zoqjMA#Y$7VlY{%{;=p?|3<*Ymv*TxI+((6Rcj_O3y&6JC*ir#LvkwO`-s<fRob8`|
zg4iab<dud-X*@Ff`0?!Id5Di@A77yO2eeAPXI@Su_0`35;SVYh;#itHoHh(saH2?L
z1Je>8c%iADNoWQ3rTY@PR&pB{Iu6F;%?1V@&W)X-W<#HzrcrFXNNA{gZUjRYR6-(Y
zSO9oC@c`0tu!V!i3Vw3~0*-ZNFx%A8<~m}AhoK&L(o2SEWOjT2nmKplx47JT9)*?l
z_eu92Nq5H+q!MSDyjVA5zQ&CRAvWpw28JhKRe}j{2-MdImS#IoeL=Ja(gyC*I6n~M
zB~Q$BZNuAQ@cJXFojL{DINEOLT_ug?g=#b1A*yNcP;HJFs)B~Zbklaol>da<h@J!k
zRq=3y$=eCv4Fz-P<QaGk30&xzef+KY=$hJtSi6SM+5-XrgNAeFxWr_52x&k_0X;i8
zEIqjz7`jQ{YNWB_(1*~mub?2^`xn>?V*Mdwl{huvD(o;ln4K3~4m-|I)NcUqc8p|e
z&9mQq5j=k>>^p8;oV!EfXfls5^?|zsvrkR~h@{K+aYBbi*Gj+ctE16kiiN;pju#9C
z2YwUq_J?EF$?rftDkJMIbnU8i^>ln=NfIf9>Ef2lm+*MFIb8UU&D%v?*aKJT#4P!X
zfS8jMOp3Y55rUW=+Xh8rf?E?aWqwLbi#o^<X6{1wf&7wRfo;nshifx@V#VAwx~CB4
zP(qjkYk@yFe49g<?toP2yKv$0OEJF_h?o%<VDP)JCNB)B_#HTe364+=kd{yK$9DWz
z=fGyPMfOejQX|*NV={y2;&)?d4-!m7RH<k(<t-?SI$WVGoKwIjuo=<}S2kI<oD&wN
zjl-`voiz)lxDwev1ZBcJkZxxFX9@;B{E;-*OktJ3qRPLj#=pAOUsmp~tnrtx_Lr`n
zH|iIe{>T9N0f~3Ve6Cz(Rh&$<zkF3#CJSY<P$o+*W2(Hj-e-Dq%Jk;6=}qsB7m+;M
zjH%MUY`MR5C0buv>R-InUtT$%sVf9cKz=}ysEG)a`^zebz%qYnxqrz@XfetF0qbFz
z>_Tr=a{33SjHG{#LP3`PG1H_kP5A;<f2n9ubt$2wffB=N3DuEW72NBg1Sw(odad%P
zn<04y>6)}iY6qUD6aEZB4TNR_RYO)Ec*%bNWQ7y5I<lmAMDQyK;E6m(^!_B%f}9ae
zF>@(&CNrhb`3fdE=R!;|&i9~O@Kq^QOVp`44!kMaFRP6y$CO_rWl2_9b}PrbIKe`4
z+!kj?o0!+u?mWmHakU?gtIW0Tr%9n$T|!P%%)u}85XV?r+ubyY-C4Wf(oEvN3yl+o
zfyC#Z&41^k=lWefdN$d7^jyZmCk}tv1|K~e_W0=e^JX7COIv&fSQhJj^jyBdXISdb
zEAy?~;A`0B%iHBM?20A2%hn8U9Nu_u3I4`C`1}1n{P{}vqLw^Nrp{{^gUQ=K))Hia
zF$vmKrJ&yu4EV1nPHP)yv54swe5vSgVA)`?eAHoe&?meBMEJ!>mLtLMK71XLw~=68
znBRxQiUeni*do4<$IK=nSP;g5bVq;lN`)p9(!esJPS-4+XQE`0L9=q6iIOE+d>SR$
zI?Yn}avdQJ22DOP$CD)*P4PSvC0Xw&G==j_9A%Yy4J4yqL1<U7@vzNg!$s5p;9N)O
z2iGDRX2>GaS`ujRYT%{4mygJw^c(3gzs2J=23c%IXaxfQwxLk?ne=Y*^E0`=$mwS`
Z_~hrj_9LYN+MJsHk<|b2il?MQ{|kg<yAuEa

diff --git a/Read_and_process_image/__pycache__/__init__.cpython-311.pyc b/Read_and_process_image/__pycache__/__init__.cpython-311.pyc
index b14b6425d13cc161dd631ad55f920492e124585f..ad3439c67b5eee0005ee8281284a74059d862d60 100644
GIT binary patch
delta 63
zcmbQoxRsG-IWI340}y2C1gB5rF;@TT3goz0#T1w1=O!j+#3v`_C8rj}1Qg|`7bWH@
Q=qkKe(*9!c-iiMH05+HvZ2$lO

delta 40
ucmdnWIFFHMIWI340}%Ym2~L^FV=VB?(A6r&FTgn_peR4RC^2_pnm+*LE)0tR

diff --git a/Training_Tools/Tools.py b/Training_Tools/Tools.py
index 94828c1..8a01197 100644
--- a/Training_Tools/Tools.py
+++ b/Training_Tools/Tools.py
@@ -1,5 +1,6 @@
 import pandas as pd
 from sklearn.preprocessing import OneHotEncoder
+from torch.nn import functional
 
 class Tool:
     def __init__(self) -> None:
@@ -19,6 +20,10 @@ class Tool:
         self.__Normal_ImageGenerator_Data_Root = ""
         self.__Comprehensive_Generator_Root = ""
 
+        self.Training_Zip = ""
+        self.Validation_Zip = ""
+        self.Testing_Zip = ""
+
         self.__Labels = []
         self.__OneHot_Encording = []
         pass
@@ -43,12 +48,17 @@ class Tool:
         self.__Normal_ImageGenerator_Data_Root = "../Dataset/Training/Normal_ImageGenerator"
         self.__Comprehensive_Generator_Root = "../Dataset/Training/Comprehensive_ImageGenerator"
 
-    def Set_OneHotEncording(self, Content):
-        Array_To_DataFrame = pd.DataFrame(Content)
-        onehotencoder = OneHotEncoder()
-        onehot = onehotencoder.fit_transform(Array_To_DataFrame).toarray()
-        
-        self.__OneHot_Encording = onehot
+    def Set_OneHotEncording(self, content, Number_Of_Classes):
+        OneHot_labels = functional.one_hot(content, Number_Of_Classes)
+        return OneHot_labels
+    
+    def Set_Zips(self, Datas, Labels, Address_Name):
+        if Address_Name == "Training":
+            self.Training_Zip = zip(Datas, Labels)
+        if Address_Name == "Validation":
+            self.Validation_Zip = zip(Datas, Labels)
+        if Address_Name == "Testing":
+            self.Testing_Zip = zip(Datas, Labels)
 
     def Get_Data_Label(self):
         '''
@@ -80,4 +90,7 @@ class Tool:
             return self.__Comprehensive_Generator_Root
         
     def Get_OneHot_Encording_Label(self):
-        return self.__OneHot_Encording
\ No newline at end of file
+        return self.__OneHot_Encording
+    
+    def Get_Zip(self):
+        return self.Training_Zip, self.Testing_Zip, self.Validation_Zip
\ No newline at end of file
diff --git a/_validation/__pycache__/ValidationTheEnterData.cpython-311.pyc b/_validation/__pycache__/ValidationTheEnterData.cpython-311.pyc
index 75327b1d4dfac21940fa9cf41bd3545c66b5fe4d..f3a8bd9d87276abd2a2e8efaf7928c0925bc2e48 100644
GIT binary patch
delta 64
zcmdnVae;$pIWI340}y2C1gB5rxvl=y1ITf)iYYG1&rM9uh)+(;OHM6{2`I`>FG|c+
S&{cS`r2WO>y&JzfF#`a78W$P>

delta 42
wcmcb>v6F*mIWI340}yO`)Ri)k=eEEv6IZJkzX0c$fTH~LqQu<Ie2h-a03u}#qyPW_

diff --git a/all_models_tools/__pycache__/__init__.cpython-311.pyc b/all_models_tools/__pycache__/__init__.cpython-311.pyc
index 1ac5718201d336b108cd7fb58f25d728ce53741e..e397ef7b9b31e2e2f3e1aef1242ed69f2db637ba 100644
GIT binary patch
delta 63
zcmbQixSo+`IWI340}v$W1gB5rF;@R-59GL5#T1w1=O!j+#3v`_C8rj}1Qg|`7bWH@
Q=qkKe(*9!c-iiL605HH6O8@`>

delta 40
ucmZ3_ID?UAIWI340}w3JI*~k)$5`N(wyRZ)Ux0H=Kv8~rQDW}IG*1BDtqf5B

diff --git a/all_models_tools/__pycache__/all_model_tools.cpython-311.pyc b/all_models_tools/__pycache__/all_model_tools.cpython-311.pyc
index 31140773e4eb0c3af3aec4732db6cfb0f209ad2f..ee90445f0fcc8d3f122da482a1b0170ad05d7041 100644
GIT binary patch
literal 3292
zcmb7GU1%HG6~3dHktKO#N%2oyCw9h8v_#!l*^nlrs_W9-G)uCM<Hp-`YfNV+nvpGg
zq;c<zomj{qN}w+IArA`)WD9~W?k)~X9@@7q^q~)ZtK_P97znhK!sewYIQ!D4o^!_<
zSy{KxGt#-|-h0luKj*vW-rq!{5dvha{bb=okp6*{76}K!{@WltA{x;+gA}+qjzc+M
z1m*&)<mdPRd6#JX1ELAr0YY}*@6nu~i4Vw}s0C(8D)<5$Q=BW!P&2DrmU1m`=+oH)
zT~@R#_s5|6i0EXF(+G?c(u4=x9M8N7T9End0-s?w`U903>$A4GvXU<@ytn~&*C@Kl
zw&stHv4MyG{v5DJq|B9x0+L7N0Mt0Y#zU<&4_GtQ(2|ixpyvD4&`xak!F>2>Hukkt
z&<(Aqc3v-Lb+>Cxr*}<DcSTJ%Y}M_`nyW=yr*8PJZrO@qT9zBpaT$sQ>qrT%VChDV
zhQKuqBf}tBIIaxz)in^VX68Y+pk^19tO|qBdDzqisutvN`O8m!@#U|-m{$!$DVQ*%
zrP!uvSnUR{o?cmZLyD3w=50kO4Rq{dy4Bu`BfkZ#LYzT)Q`m|+$48)$oc<vwBF^za
zR{O><6g|$+DJWu_@a=<5XAFYSz+#Yty@nW&B~)Qlkzb1lsrvC~>0k+3fik}hUUvML
zAR1>`MBDzrw;#b0*e$b|(I1w{eQwFO)VNX<<``(l7wr7#(o=SKdu)t#+wuvZ9qPCu
zMo-AEgl(Adjz4P-*XNIPSa+ErhK4l((9_lOfNN;2;}A0czkLupv5IGI`UOs@bi5<F
z@&N!daw#O6tF~<B<o6kHQz0sWz8kaDHC@Rr>e=NLGhehB0x9kX1EediP;6xgg*_yn
z3A6|8LTjqw0a|3xbHmIL9EC+6zcj`)6#y<E^a$j6?I%FX`ZF*<UUzJ3verH73@10Q
zZcRH<d{2^hCAm8I>7{3>x-`*{CTgvp6YYO2eBASC-_xP`;KkkO#ai?tSZq#hO+Eg8
zT^dHgiS=%kn}m!0n0p@YuMPa@nebZ)3i$m&`(weA>hU{`_?=q(4%2>Yf4s39J6DUH
zbH?6yD0~!q2w#&Y@o{Ks2Fb`9|7&W_iIdPhwFw_hz?wRe6Ye5HW&aakkBIGqh>#^;
zE(fYUq(RN6ZIE{S8XOMLYA4`M7}8N2)5!s0YFlcjufP)lK+S1b3S}N39ef?qEs&<f
z>C%Zis*%?q#hFDJGnlODS*oj+uE}RheEM>(B+BFClX9sqm6qjMbq%t!%rGI_rVRN}
zv`v~_gj5Geb0e0m+Pb3Uv$mJ<`eBj`igW;^PVCS=XeteNKsx;zV84l01z;8OJQUrE
zJiN5mdwRF`biFs(=uKAF>!Ayc(1lv)0%B_dpb);#J(m)-V`~6~9w@xup9X!gwrqQH
zUAoth?$uhqCXcqB2Nv7u6i+cTc-Ii#1d38r3%a7X5d|~ys)4ekC_h_O4X=mB!GI1S
z!^M3@m?w5pD29l!V<1@X0jv05GY}NzCh27Hs5smtJ{#dhoZr#h8wZET`q>I9J|uq*
z0PKWhC*U@Cgc0W04FJu5$=YAfjr`GI)&jPl5eczz76;S@YSquWtv)O}XVnQ@oSnua
zKkU8C9ib(lR{6D8%!X)V$~quT)PhgDI*-{#(8d5B(2{@1mhj?cR6TT;wAx@KeA$Gx
zh}Nw|o=V%WvK@bQ8(n3w94vR`c*x_?>C#9#eg3v?ts3@uhIM;USJ?Z39jfyssa-iM
zmv}ifUg}QY)AI|9wv}GI#31>8D$ssaEcAW;?%zv&|N7nwOe`$@yL=&EOo?tVr)J@J
zwsl<rM{$GDtr#>VQ3Q*dxUs4&=(kPNR&J`x`qeyjdr(l|xa$hGyS!ynHw1Y?xAO(v
z6;Q&Xuc_-Euuq~R(OiLUS2Cz#!KH&Igu02_AY}Ea+qa4{2AHX<sgN6Ct7YfO74t<+
zzwd^balQm50=_8Su&1NJbUgJOv|4VMJ+bhG<%TJ94NI{QWB95<@@}RuyMOMg*=3r6
zPJH}We*p%Wr2F_UC$}amSN|4&>sh8A|8C{FBZT&Zp<Q9<@we+jvLPgELefd}?<LOc
zCeBoU`n&8GXMaDwm$|u{xmnNLYGiKJFV8eC&(xDMwb>8q$q)WIuhc%&>+`wBe6F6%
z)$(ig<XV%Eb?)uJOG1vl6L`gV<;GVbk{GGdXWy#DuhipL8u2TY>;H`Rf7IQGpWTaR
zcH<dmaMT%1I;YP#W2c={XPnWoW|Snp^OAHW23{enTyGvD(!j&@YO2=#W=(jLoqotr
zC#OtRQ&zlZc^bE?Re`L{VolG%4p0o17dG4(I))6dGq!R64o$$uL+B>y1hnDJWl>;|
tAUVz<muj6qhr}xEbBI)7UvuOHhnebV{*tx7I@~+a$DM1EuQN8M{{U_C_k{oe

literal 4519
zcmb7HO>7(25q`_vC70BaA}v|A9NDH7CpK$ag5@eng(|N61KF}IE4EXoTO^w`cO_BU
z<tlHNZp2arQRv_RHlUl;n@+9`355?0;A0Ox<j?}U0fQ0?7;sVKAfPB@IOxHbcIHu}
zL{V+i$KBz3GxPpt=6!GWxAAz4K>6qEW#f-f`x`duMcAyo`Wh+^iB5FRBo?P|R!9kP
z*bbSzB`AU=Dxwuu!i>k85ld1eE2>1Tm=d$vls3i}OxcPnaaI>$oP?6F+Ld-IsU)qG
zlCt(FdzdV2c37QCC#y%yy;hgfW$jb;v9@IHSN6lalG&|vhsY?=ql-k3Jqi)B3U93_
z2lTc@qP$^rPI9`u$SFN~9NGu<1kgizJJ7?rbdzL~>)0U^3L+P%J!co(AU2Y9b9Xg2
zXXi6~5WlEVbN;4l&&}oXQ$gy6p%=2o)f?CI6Q<@GS|Mn^X6uG|e%i>+&e^%Vn-KzG
zMAxh!a^u>~IT$pEKy!*}dcfZty^!Gok@d)kLD#5h-_gvGy9U*!j0xK|VUB?dfq22l
zI|dA3IMdpk5lGhxrkk5H=L25Tbr?aQhFO3`r)CQJlySqhUDeWNjSD#%B(S17+Fe7X
zz#q?UT7>=8?epp?G#?U!D4b4|kRI0gMM2>WK^GRuBap2&dMTnI&V=htU2K)>A+Td4
zh@YR<@>7O-)79L96NIVZ7HEDv%hmhBKhDp?`>GQNL=#scD%5I4CooQEoTAbbE|9Nk
z^9EI~7u>mmi+gqqbJD@l(y!m?xAJbk(LX+MzJG$+U|g$TGfmY(S2(H*mUOllwRYd!
zd?2Z6E}wH%wb<3P+P*qJ0mD1(K+2@rbMT%}k-Dprd{1}<ERwC84;wqSQfk@C6K<S#
zgJLE`-+(`Q0EmOV(rMZQe?b_c4YpS7Y+g!@n*xnjNg!qNqUShUh$nT!Lk|J-|62)7
zLznGq3FOTwf2x@UWBa1$5m4QIMd&ixiv(8yyV?hT<KTEOYF@JpRSjY)dZ1upT~^i4
z3YuAy(1V~GNO&3y>dv!Sg0N$_u0fyF24ps&T{u3HIK0j$K+28ZMkpl4H%K!HlW<}q
z#EF>=0^}tR&Bc!x<AvC^_;ItSV>^R&V0&Q`Hknj_iCLV!1C@v1Yj;}|m$*mJT5WJj
zq`AEiD#3|hM$d2*hXEWBb9wM}Am|7P+~*Fm)AdMbgHx@d1bk2XL9!oS2ZIE{P}=(-
zvLt*FzaL*Re5t1*^^`}dZF2ciYa}$5{T_5}UH01dS~wg*IBhX)3k5lf17~=868^R%
zfLeXPXx>T!w^Bz?&^QAcxDf{f3wLwkL2M=FNhf^iL`6FB_@XZjRHT8XEj26)T@ZPg
zCyoO5At{j=VAP7v;qj`K1rB_KdYh#g!&*HA+G1UokP@edif}Y_eobh;M@pd`zU5nb
ziY<NkUHU|}W&^Yhm$=CtTBRNB1)M3m)obUDO8ivI+Dbek@%Z|`psXW!&^<ulo)8Ow
zdJa7etsvs$rYt+BucMja5)#=RL-D{-gg=V#N3nNCsChh+1!T_~X26e*Oa$T`)6UL1
z7!nY`n^@Ma!BNiRovAW0eILc}I?F)7?}?84v&+5r?egf0SnoHnv>)rO#Clf-{aB_F
z%aqT*kcS^!Tt2pP<2R?CPI~gNFArDb;b&Q2K39>?l`mH1A(TIjdh(Dj4^`x$XM?^x
zQjteccJ#sM(g(}j^1DmdefelbKDzRuFQ2H$C(0LJ$XyREeKB@_Y*}0`Ed9!t->%4S
z1G_C*zWQHI&@)&zgd3-GvxwPXkR}KqYdeh}0$skLyDbz;yXy*@x~Zl>B)Yg3ZVo|!
z6HbpbM&1@kyN-;Pu~f{*&^K6EhL9c}PM;o1V_MQG7#K=#W-6_+Gec>}Sz7r&WP@Q+
z!({YFU?M<`xM^rK4@rzlAvsz{%dMj+rvCgVEPp=>*Tzqv9Y~gKLA(^Kbp|a0m4Hhz
z+-VyUOh^GWLnaMGwu4=uW*<V0J?G}ET+yJ8I$6jAX4!c<3Y|~E>)1fbWFtYO_!9T)
zV!8NSPOj|n<fFA9xyMx_a$V4|a?X>F`SP)fe5^^(z7l%uc=8!vK2wp;G>LZf*&UQ|
zXAHq^Fa*P#rwFWtP*Xtd2q&%y&GgXrAaNVn7$GxgK2}z?;aQzWcI0x+xP+_Fd8FG%
z-+8++EdeQDAvkoE9S59Yfp-x!HcN@uMMO3|vL-!(wX8PQvmm$#DVqhfqIztt?Ge~y
zwQ)=qL>DujEge~p>j^!gC)U~@!G2d8dtL~agpybaPw@+pQUroMIbKZn_4UKUl3sV(
zQ12KHB*d5t_TRGSa@qc(yu~`6F7oNjsUqJuJs3!LHB&Wh$GH^<TEVp;d5>5&fTazY
zG#ig~wDXXAr&#+JMs8}_byUU@aFX>gSa$6M@bn7QA<==)2P2<_6dta?JGNs?b^dPj
zuf@)PoTVeEj5f@KYs%qCxN2=mCs6PKl5rqG*vjQqlV+l{19_Nz1-#?ZK+@r8LC_nJ
zOh7FVU0c`YDdGY}n^QDZMharg04z1b>_ay`527`e4xb4%>156{oVun3A22}_4zM}z
z!1qoNrR-Y=77-+}kX_%=0QOZg=eQ2WG(#64wPZGm8Y-&LPeF+A?|cOWpg%eARPmE%
z%a@-EdzW&)kgf=6Pe?yc9bQR%+41E^zyJ6zgO|P@yyOp#RR+iW(^o2|ulVUJUm1S-
z<~O$$@8_TTx3$V`%};Bdp7YZ=Z`Sowg-WVW9{V<xUODho@g{D0GkLG@OD|ROQ>99(
z1ft}@m6>M)fB4KxUiOohE6K~{%ikuCRC`ZWkGxfV^JumA&FbN|sz-V^;*nI>O9JFO
zB<0H+d&s^cE1AbX^Y*@59<R1_dqOSk&ZGl*7QWS-J`=x99G2X%<S1yb@f&v9as~5D
zkgUnK;A$J<A=ok4tB=my<WL#;yWGnZk4liLjkuMkIqbF%dal}<u0pDvH5{i8JzQ@9
zE^h@B{SXHU*-kCz#;lViYwYG`5Xu}$e}Dqa_?e$yOFxFiL5BZ6odzDJ5Y8t+H+YWY
zs^p}%+fyaiy#M@E$$Q@adaC4**Zfq;xcB{@D(Ur_pDG#ic6+L%z07`95-qdeh7{%C
w%c7OMMBaB>g02vUpC)x8k>e+~HMa@cxYPz|kOS@5qd|H*xC0xcl`#AN8;eXi@c;k-

diff --git a/all_models_tools/all_model_tools.py b/all_models_tools/all_model_tools.py
index 924c54f..6ae048e 100644
--- a/all_models_tools/all_model_tools.py
+++ b/all_models_tools/all_model_tools.py
@@ -1,40 +1,68 @@
-from keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint
-from keras.layers import GlobalAveragePooling2D, Dense, Reshape, Multiply
 from Load_process.file_processing import Process_File
 import datetime
+import torch
 
-def attention_block(input):
-    channel = input.shape[-1]
+# def attention_block(input):
+#     channel = input.shape[-1]
 
-    GAP = GlobalAveragePooling2D()(input)
+#     GAP = GlobalAveragePooling2D()(input)
 
-    block = Dense(units = channel // 16, activation = "relu")(GAP)
-    block = Dense(units = channel, activation = "sigmoid")(block)
-    block = Reshape((1, 1, channel))(block)
+#     block = Dense(units = channel // 16, activation = "relu")(GAP)
+#     block = Dense(units = channel, activation = "sigmoid")(block)
+#     block = Reshape((1, 1, channel))(block)
 
-    block = Multiply()([input, block])
+#     block = Multiply()([input, block])
 
-    return block
+#     return block
 
-def call_back(model_name, index): 
+class EarlyStopping:
+    def __init__(self, patience=74, verbose=False, delta=0):
+        self.patience = patience
+        self.verbose = verbose
+        self.delta = delta
+        self.counter = 0
+        self.best_loss = None
+        self.early_stop = False
+
+    def __call__(self, val_loss, model, save_path):
+        if self.best_loss is None:
+            self.best_loss = val_loss
+            self.save_checkpoint(val_loss, model, save_path)
+        elif val_loss > self.best_loss + self.delta:
+            self.counter += 1
+            if self.verbose:
+                print(f"EarlyStopping counter: {self.counter} out of {self.patience}")
+            if self.counter >= self.patience:
+                self.early_stop = True
+        else:
+            self.best_loss = val_loss
+            self.save_checkpoint(val_loss, model, save_path)
+            self.counter = 0
+
+    def save_checkpoint(self, val_loss, model, save_path):
+        torch.save(model.state_dict(), save_path)
+        if self.verbose:
+            print(f"Validation loss decreased ({self.best_loss:.6f} --> {val_loss:.6f}).  Saving model to {save_path}")
+
+
+def call_back(model_name, index, optimizer): 
     File = Process_File()
 
     model_dir = '../Result/save_the_best_model/' + model_name
     File.JudgeRoot_MakeDir(model_dir)
     modelfiles = File.Make_Save_Root('best_model( ' + str(datetime.date.today()) + " )-" +  str(index) + ".weights.h5", model_dir)
     
-    model_mckp = ModelCheckpoint(modelfiles, monitor='val_loss', save_best_only=True, save_weights_only = True, mode='auto')
+    # model_mckp = ModelCheckpoint(modelfiles, monitor='val_loss', save_best_only=True, save_weights_only = True, mode='auto')
 
-    earlystop = EarlyStopping(monitor='val_loss', patience=74, verbose=1) # 提早停止
+    earlystop = EarlyStopping(patience=74, verbose=True) # 提早停止
 
-    reduce_lr = ReduceLROnPlateau(
-                        monitor = 'val_loss',
+    reduce_lr = torch.optim.lr_scheduler.ReduceLROnPlateau(
+                        optimizer,
                         factor = 0.94,           # 學習率降低的量。 new_lr = lr * factor
                         patience = 2,                # 沒有改進的時期數，之後學習率將降低
                         verbose = 0,
-                        mode = 'auto',
+                        mode = 'min',
                         min_lr = 0                   # 學習率下限
                     )
 
-    callbacks_list = [model_mckp, earlystop, reduce_lr]
-    return callbacks_list
\ No newline at end of file
+    return modelfiles, earlystop, reduce_lr
\ No newline at end of file
diff --git a/draw_tools/__pycache__/Grad_cam.cpython-311.pyc b/draw_tools/__pycache__/Grad_cam.cpython-311.pyc
new file mode 100644
index 0000000000000000000000000000000000000000..ece66234c2a3f0a5ac991e9b64c6b06682a05828
GIT binary patch
literal 5455
zcma)AYitu&7QW+|vB!_R94DbE2?2tL3%G;|%EQo>cUuTc0@~IKJ3DwL#$o*E%$PuG
zck_d4-6~R{N=0_H(pp-?rh-cR<VTU(s+IcF!%l^DB_yO(rSdNotyEPj?VdBXCz%Aa
zdp$XGALpKP=H7e0b8i08(h?w0CN__y-ftu1U)X3Cu2FeXgvxay6PZy-hDk6C);+2x
z!zNg}&Z=C7Pk8M*r}`2;80S<W;fFV`1~S1!$R6{mEt%FtYlE*X(Qfl~Bsx6g2$6kP
zh%DUn5ONFt>Ln4DxiJ#;KgMR1p`qVsxul}&;^DNaK$}0FlNA*z-u&gfnlnbA>N_VT
zFDO|VhtnA;r6e1|>{nCtldaIaP85=0WRmd6g3Mmw60FQ!Aqh_Af%39n_FiEUUNj6o
z$peN2sBlD+Wbwdj$EjCSjDnUON%BkNqj(nnp5$%uMYc#pXSm2gi)l8C478f&kfH}#
zIMXggQfaP!v&(+p1bWNqff<f&nT(N7nUScMwkenLik8kOSwqZ98HEOpNtYE(9Kkw&
zOgg8iwB<xr5s&5!vEJ-Hkjq|FG(&85pUR2ptU+7Hq>GBZ*7ocuPdQyt$2B-9g!&rH
zwC&KMLR6x)!my5(4Pg0s7t{{!KBF7CjFdbtCZ%jr(ayk5rZg!tuw~$f8-M!Y@AuBg
znlvdIxtyw>vDY{$Wn%fuR1m?Gv>}R99nGbWHM>Hn2*pp6+L~CIvxL>P-oY|A(^l*1
zg;$`~2Q5qJt@W>a(7)?`|1PtCkJY~i`<{O!gv#6#m`svO*IIxMdk6mwmFvWC_KOh1
z*&jlR>`fTC<&3&p2g%vrmpH@O9q=$`E}RMOIaF&GnYrz8pZOC1tlcm2MP?bV7kHOw
zX-WnAyly>3GU08GF4Yg`j=KpGZaGpIG9%(?in1db9?wr1%1M{fdeldGP0FSe%Bo71
zda`*Mgrg>9Wl>INbn1iclws3o2t83euSiBl%4=Z|5%p=E@S=PM-4)z2D=U|%*Y-}G
zGHJ?7nkHSQzQ!co2i*4I85q#Iab_sraQuvv&Q5hXCt+!K7clB2Ak*YwNB1ZC+>UGS
z&c17QthYMWFA&y$imB7`YY!vcvm=!q^J{PHx-(pjY&Rp@t;qKBp;~wEb@gMlvVDGo
z*}c{3-da9V3$;H8_1zEk&7~?A=cOB$?+jYo4t*h6+m3#_p?Y%E42@Z#v8w&pCrVod
z79^c>0SyNe2B85s%3L9^8&I#8gT(FCu=1V}mjK)vT)!tffLtdjn0~gq#4J6U6OKYS
zBCdmZ%`?*j&;UZUewa?Zl9tLym!cl(iEXt#Sw{mK3;k;#uBLS(nakw&O}*qS^V9Jm
z7FTmgN!9m3r8x<8=<9$?lWN<m^`~;^6VI%FZiCrAV6_i8-S$dF8Lb_-;e}tQ<7d>j
zT<d`9)SCZcLLW{-Zvq6c)y<!>w!8u=%(ho8=n!75?*|7g&4PYldw3Z>-0k5-&*DO+
zn4+hGOS7WD?m6D<UhD}+4@h$AspTHd8No%Ni+<@o2ir-B_YjB~%Qo~3*i`bCd_{J;
z*Chv?#x6yl%q${|+rknHqY%H}_Y~Q2&mvbzz+RcS@r5DT;CeDlbZ*jfiG7<)G7Nc}
zFz_q#6-3Gvr6?4=ZW$%N%SwN9O3^QSZnMjH14Zvr+e==+v4G4Oc9v3~IpNfo+CK%W
zT?)E1Icr=DE?WaPA4U<o6oc|Ie0Aw<Y4TMtz^y3DwM>EjQm`1TV6<FO+<Of^&)3m5
z%G#j&7#-m;o<6M?NG#f};c)`Y%;pP*PW>sx5LJxpcC@A5T)}_=>M_P?2z@3E;s2Bb
ziNI!@CZrl&G^i<ZA*qNNMaoiNR+&U-r#}6if}%_*)CbUz&J;3~%Sgs~8fbD$mnfet
zWb&7zp$3${j;WQl4y6b3zy?toK|O$)#5&G%1{grQRY^C*BxD$MgT;oaJkHW-OA^L)
zF?n9fW-&J4d>!x}5&;lux3rC*MaOMw#02|g2L^2bdZ*ftpi)T6V#p6MXe~}i!5U7J
z@4I?yE#a9n<&nA5e-r;A&Y!w1ntgjMd=7r0e63bjkNwDcx&uEEAmzZ%0n*w&BbqJ!
zR!jc^;rzpu!P@2>^KaZKn49-noA;WbXf?F1a;Vn!T(xcO!@f0teY;YaKY8Pl*|*i|
z+j=!nTfMeY|0OD0tkqk~#~1t{#+GH1A5rR0NEZ8^R{Q0bU!K;z&`LVGuD5*LGMA}t
zIbgOQwAv4r`JcUHRqyP373EH}+7mZ>;#N<*99r;_bvy4e*4o`wp}!U$nvdUQ?~Q(*
zxZ7rJI%<ZGmVFPyk=YmKertv|Sm6z2ABgGf!j6h$hKH>1&}Z9jzx-g+-us*O-c6aC
z4qKZJo8iM{U#+eCqtb_^5B@mwN06*zl>f5z<^3J(R~_8`F5g$H7@$dn+GcTJ_b7m;
z8+P5!Q1UKKv?d$~E;7sAv&0k`7q0`$%RzGrqB2BG2YetQa}pl6b{k%OCBbmq9e5X^
z)gvzlbN@3qnH8LFTOIh6*&;Iq)+_@l#wFLWyy!O^zyix0FkA+t{-RKE0I<0Q_y*YJ
zZu4$#fb(Z@Gcw$NZVNYIWp6pF3l})Qu^2muff&^&6An23Kfvey<|&XN2aqn6zC6{&
z@e%}z3MLLKN-^M;3E$?eU@ozuKp4kkYZp1VLM!K%W81cpi+>7N2zkK{IJt{g*;2rD
z`Y_wE%SwA)e32%f+~U@RmuLtZt&OnZd5n3c_ByaX{tplh@3(jbmel-tiLw)lk$TB4
z|9k~<KQ4*6`J2EqH4uxvII8Fc)p+s1$cU)fRWUaXH3%cw)X)?=5Pk7iX3G>8%V$%R
zl~gs_rQu<)GfHg_)F=yqgR*(mpn_gFhc^J7vdN3v@S3crA+qoVxKZw;Lh_<<AgAUu
z+H&B;u@j@>{v)H?Mvv^r+kk#PH%VFD(5N6whGM`SmU395ypfaPe$eqoK~5>7Ik;XN
zmo6v=(;98XifG?cv0p>4g>VM<*70<T@<}zPE771Gv+OWJ*>omlr@h*GWaUq#4HeYj
zKNKj7b3E`N0x*erjDUU=C2AmAM|WBhfa>uFBhAwL$daqiUKrA|b`W~l_WTF$&AeCU
zR={i7_ef~_WZUeHxsBIenSI3+)&T|!q_cZ=^R?J)>;qq!EssJ+q-S>bT(Yv;?2K8R
zv9j<mTu12%GdygChwC^#H1|@a=g)i1aMTJ%%f9bHO1vU`w%&SSkJ%ZwI^$&l0)jvI
zk^e*g)uGDTvfmU2En%=K3_cP9cJIatKR<^3!<H~y6^0*0cHCp{9sT_HgZPR2@e^kJ
zq!mAD?i{stj+&9t>gj|TNql$q4^?U0Je#u4rp!pH?5_!-2SWdSp?~hODMT$HS{0(T
z$f^gCjrSuP=Q(RAepfe#_J1?>ZP%BHuM_6biR$QSbLjMUZ~d-%Ml#<zXT5dK96DE3
zE|^0Xs+o&sWYUUEmi^xg!I|LI*G-|{68a(RJcPFY>IPF-WeKaQ!Ya^P=IYl9kA}!N
zq9ImB1^h1p{x3unX+Xs6q@ZFwB#Q48B(*-G^#H3jgaqNj&MY+CUTqT+yw=sX9HCf4
zf(JrZflNEU1&@z!Ss-o__F|ty+8O>exQZ^JLNC8&fw;-eH8^P{sT+doqn&j1u^c2m
zjZ1KB9L{jF0gh|CC=FduG)a%y8JS*BA_I`yp@*vJa~eiL?Ie;hB>Rxq_e9>Fr(x8l
z%$u~cE(UV8hBj1yVUY$5C9CJOaWyxoZO7>eB=#b?7w%UfUSIirng%T37)1XD$O6kS
zOpR=>x}O^9uezTa8LYaW8VO9>zZwZwo6plGZm1I1QzJdq=2IhW)AnzHJIOGx=uf9V
N+WkL0W7OW3{{na&!~y^S

literal 0
HcmV?d00001

diff --git a/draw_tools/__pycache__/__init__.cpython-311.pyc b/draw_tools/__pycache__/__init__.cpython-311.pyc
index 747e1cf244c17bc14498330912ce3afea35312a8..20957044eb5c62c0194d1041c6e303f6e887f1fe 100644
GIT binary patch
delta 63
zcmbQlxRQ})IWI340}y2C1gB5rF;@R-3FNp~#T1w1=O!j+#3v`_C8rj}1Qg|`7bWH@
Q=qkKe(*9!c-iiLs04~rKJpcdz

delta 40
ucmZ3<IEj&GIWI340}x!$I*~k)$5`N(s;gCuUx0H=Kv8~rQDW}IG-m+d)eK_*

diff --git a/draw_tools/__pycache__/draw.cpython-311.pyc b/draw_tools/__pycache__/draw.cpython-311.pyc
index f8d3ce6d630a23edac22b12c4957afc0ebfd5ca5..497eb21eb567a7524844ed39a8fda2e1555800a5 100644
GIT binary patch
literal 5585
zcmcgwUu+Y}8DHD8>pzJT8xnFPKo*GMtdc-l2+0Y7I5^<)4~~Sibt8Nm@5XUhdwsKO
zNGwN+PDs#-Q*%<QM)%~qqckUY=*vkx38X&qxErmMXeA`1>ZHqCBy=xPANtL%cjI+P
z3T>~G^|#;7d^6uSGrPavH~C9A9HO8+|2mQGuA``bkO#YXE6S^zp!|vADc&PeSx>_2
zA-Y%eC6IZi6SR3}vi^j>(iX@DQ(%`3CF;x?pIDy_C&HkkyHuiqr$ILI49E!Y9iig>
zHR3GpF_`>pUX+w!Q0uQ@DJ95q@=`_=QgkI6a`VH_2ftTeArhJ5A(d8n1h>cb12so0
z#I91N?HR#Lo_HTYug6YL59P9SdhtFU@xDj!{Mf0hv4t3{qQDB;h4+;x-dBXjN8t6b
z<GJMF>Akj5P1}DtuI~-UWo-96v*Q|&RW=Ln-_b7M*ai4t+RKNY)Vb$eGq0dyU%%tH
z&U%}Bfc}Sj;QfjAz{|h=9)x#XgP(p68fyLXjY_k<LlmX#+0>`&ZM__}wd%Y3+jEag
zz63r(<Q!9?%MI22Y`wJQeFV!!yS4gW9bpN%J@HZE#bx1hTbQxzA`(2hzRo$n3k4#!
z&(h9cK-9M5ce_0i>;rUb8jgF{$Y5&(E57j<{>e;Q&J=|;f<J2n954|^;Cw1oz+7t9
zpoP4Ynlk87%w=*$Xp9pxJf~!&TrpJ0=LO7)!lVKq=@X??v0+H!1(9_+BT<s&V#ul;
zBs$yGb^N*@7ewW_%-t4}$`t&_(8)}eOAGI^8(ucP;bAvWS3Z|Ee4Hr8cVQB(;meDP
z!9c(TEEs;dFhTIyK&GTImrDx<Gb?fvf@m-^mAfdUg&c3tN=6X{!zU})2tfFPlF14N
zA{w+wJU44Z`wM(pxGqUba)`SnT+Cpjfhfrl5+<n!pP+(&gHFM`gm}PUJ|K%D8w{D7
zAR9q@=AcdjStN5D_z7jns#rQA9AX5m8XlZjjl@`%34pPc^5ZFZxcJ_<tVmfdHI+<p
zxs-s%fr`?Y%d(y94-fDC@b$BC9&>k+iX@4$sdnXOje4>q$*Bx<#k2SXj6(*N{|S3M
zPrayX`R&;UXXh_|kNkK07jCQQfQAm}=m01$plV>jvv6?XyoUDZXpf5akh&|M-B{>e
zm|Xni^Jy)7P!At0_beTJ*r%aR9d)Xx^G$8~?-w=Hs-spFwUR#l_wQ(^MMo_vYB8S|
z{EPK!V{GY+ih4EFtD{~}NXv(61BBVGqjnXwlgQsJ%)X_M9-h(A2_2nK(TUfGxz(os
z{%zOv-y+QWqK4XZ)TW}gmyw@;JM!$Er~6m>2ABH=wZ37!Z&*8ZRX=rAi(FN&k7<#y
z`43l7U<I`<qt-=O&X|s3DvGT}n^&S8%h8TMk1Vx*nS7Z1tE|P|dp7d-z2Dt@dQ*!H
zJfGBJqv}{vi+-#}Kc4R=6B}3vIj+qWyDw;{Nk>hpeZ7L0fJ!-J1`mV)$@8l(;G^F$
zeet>Ah(Lr{wx%z0NO7owqNbgcq!t|K=n^Vdr&9KikgT|&YmQN>Kz`a89nu;Ee>BhV
z{u1*f@CaHSJFNhCLB)YXfZ<(^18Uk#HOX_iDpA^O|F`rJB9J}jFZs$25pDJcFy8{w
zU|BVXP21`SOCI;U`8wjoWl@tVl6JnHNbm@IvYFh?1xn28sm(QBt;aAK%N>OWsOk08
z$8RBQuylwD2677DxMfVbT(vijAyo?0TAH*KA+$xZMNIY=6Es_sK#Uq5!1g;+8Gv#B
zwF1rqh{hR%kv}PLz!5Z;%K>crCbNnW6f-#?C5aL?P%6vi4KFvd(Y{GW79eLi1!ra~
zDq`PdF?ne=Je{}+?!_pZl5&#;8Mu$N42%1I)p-pId?ux^CgHHsBnynjnxhr>{<?D%
z7lhX|RVWR&z_<o;$)YqLmrfgv*24uZcbk*XLq#3O^GTcRd~!}U{8Iv_z!Y#h^u{a^
z2Z=aD#9<;jKo~(;P?8q!g@{_AL{qQOWPKiIa!S&o)OgTftZCt&k^Xch#|ty~7}0%%
z^%$Ta2+HuDpTV8PO&186V2I6GEIv-G>rGCwqr3`LBs96p%|AafR9*BzXPHq$N43yV
zJ#=*b!i#W353}XnS~#YMWAm3*gY~~}|6TjyZY{W95AL6T-$FX3+_a?p?HA84tK)(y
z=2RqUNYar6%F9Uiw-=rTo`zR?2bOyWwB8}TcSt)qte+g#BE#xOH?+u&w?y6M?-EZF
zTI}-kBU<d5di^(AbX<>)&-bqfDB6E-@YBIR?p{cI-u{Pn^>B|C?a|O_9i3Lu=~cws
z8~Sue-E~4k-8$-4QMXgosi7_%b*ZRpouTNu71XkfS{4Dp59{c#YG1e;*3E_3<RdT<
z`vH9K{|RElvCqSMpVS;A=MbQAH8X%4tMD~RE#LtvJE(43kvmV<e!JTu{MPguO_Q0K
zrR-q1T|@4?pAV#K@Q=hY?Hu~vrXO&_?#<B^@_ARvw8IFm$M!em26{WhFKt~V-!@2H
z6=A6cl`Gs(ZBer}Rop<=a08<d;IQZGAokWnkIT-%Yktc?(ZuM6;ZaDX3wT0`dvPC(
zj?aK7#;-vxs$_C$_JRYRLng|yfTU18B5)HD&V@q71}-Pxfdd5`F$#jr79$mC75GY(
zL0{rT8S-|5s`xDNK>!+`BZBN9Cg(*PTQ8D|4~V!##APC`fH3G8ZYCoe4HiD8Zkfm}
z8}uxx-WgOeIfUXE5l2A8{f2e6Ne+`dpRp!}2S~e#o8HN4s3b3T1u89|JmRbgoH<P9
zktUO)cckP6k(1@jWF}>vv~U7C_QFr@a6{<#!TJYn3uEPzOKCOOqXm2PV9)&f1W`;}
zQln=zbWTU-RCLZko<k1CI8_}^=4p1ldu0W+Eu*%@UzW$zeaD<EAMCel`ahzsmHcm?
z>%K&R`HLoS#-nh!;uLZi5;GH+z?->$aE`OyIc&ZnEYus(O~WpW4lQnOud1KmQn%nR
zDR)(rB;o6TU2D`;Rpawy(0zjv$0zgFKM!G8I1g4H;Keu3wC3Rq6NWVR_#4U$+<ZAU
z0%stLD2T#YoPu`PcuM{X#JbPp@vKrEs{2}{npOK+r5fkWzg6mx>b_Q~PIa4Wl{%%m
euXXf}r+J-n#K|U4$2#SRW6hr6I<-}pbNCk&{Hopn

literal 4041
zcmcInO>7&-72f5}F8}`|TY@dy-1wh$MMXAh8)<6TQKM9p?YfQO)QlZ!#a&TUmRw?X
zsYnD>2tj}dMS-Y5i~7(53$&^O`Oss6ra&)2b4Y+)V*wKj2nbN*pql{Sbn2U3?()a7
zDg@|gIQ!<!oA>6unfc!6Z_#Log7T+p{}6`*6!jm{af`1_Jf8>RkCaG>v`m%hoR227
zPiAt+{<1mN{&G3a{`$*-T%auzEQbmpUk>LYHiwa;<ybCOj_2a#L@oh5^bwUyiY&AP
zqVH`g&21C2Y1-nd%T-y?Zl@V5u}~9>Qhu%^OL=Ljq-s{6EW9t}#S*q+Hl+%SQXVUc
zR$$tzCLa!dr2qUhA|7ZoVLD~Oqec3OZv%Rp9*6FgbmKsYjEF>L12k@WR{CT;euG*7
zn@$ngBQ8npYiaM<+yQ5^t`X6{uN4v%yNH2(<@}ypPz)7)V)#j917fl1#iP%6I6qDi
zqx<Uh=G*Z^-y@!}14d-;c*YMnyA#jYzW)6G<C*BUFDA7?H`)}X9qP~vVV7=3T&nxK
z0i8{c(W|!(d~txZD~vYc2~iEES5A#<$6r1Qry|uE?Otc2@1BW+#8!_4-D}BS(IL{p
z8hXy&3hKGd3D;r+^=jxlqe)jzJlroQQh+`~x6{6DqHLSo-cM=EKUXTMrMk3D&|#Y#
zT`M3I3N<VgmaWL5Am_ba;D*a*rK(bxx7fQ_C{?V`JwYysf>u&0^-!%^l`t<$a~eP#
zBP)e^?3N-*GVe`BvZAV1fT$)^J>-nvB=mSDGX~Y><TawbR4L|5WuYjI*$}ABOAg;*
zp67Rce0rDWcTuKVDO!vm%jp4&smhwgfoC-=S$?%POMuowcI%%9d{ywW#VyOitR!39
zQu`}QMX4fMtX9%w$zoIuTLExO(n@8?LWF`bMHH4Xi5(`_2a~%2766^Wn#C5tBPq>V
z(YBF1cv+LIppEPm0rZ5*BpP{}4mvbJd52BJRFx>^N!fkn&DS2>zBBRQ4pxd-C_fOf
z@IhWv6j`;&Ocl0>IF)ak!>_{BApF#S!*;GxTgd-Wc70JtCk%ALL??hCtm$=n{gw3_
z1{yNakdB53?`LZd&FIPIYi2ZKpfMAT>1eErf6a_uG|(jzUDDAdTkTriKu1k<R7Xdh
z{s#scHqo$-hVA~k|4~GbpLzVcj;<Q$s)?=wK}M$Z7%V4cqLhwO-|fGg%VzWk2D)IP
z3p%<$CckW%*|o(Ucd}~-Ue20qpb-;|=xF5I#Kf0{Z_a-``YikKY4)L!&70Z0F*##S
z&KQXqy-+j~#nq{;@W`VnBYf5jpWUHo_SjY+v3|q|jF^EDJutGBO#PK<X8#=jOZ@Mr
z4E{&|WImhvJoZJ*;3vNtGWehCzsMQMcg^Ixt6AdE^m@oMZ%1cMGSDFt9n#(U9QKpi
zJMI4(?(4zRK7gJF`4qL#NgjRhw}Ki>(@XBex?A9J(DRU3VKonqL8F%_8u089*#`TB
z+W<YAo-TL}`L(#K4WQA3HJDA#tzDY-lg;h9-~S2h0Rp2doHp#v08r(6*9*Yj$&D|J
z0+e<oh3<7W`e0iO5?ehIeb7(F#Sm#>4ZXnbZ?Mf+XN?`IM_<3iFwxcP4NZZju4r~N
ziIF`dyJFoX`NC7Cftp?t>BxZ}FFd;Jgc2hOg={<~`-jQaZ7a~+wfuK#xC(d%?^{-=
zR0seU%0kuh2}|{h-8es?ROV_bWRHAV&~RyK%z;Ix<KR4hp2uWgcj-|~cCeo03+^o5
z5v(8k)#x28iKT+Z+knn1b3CLA-quz>a)0zLu1R|gr3dgaP-$_KPQGA-V$q5_oe80`
zD5y7p5f(%-?`9H4t*DlNUJ^8L2%m=8_zY>zlI9#~$dT{L5J6Sa@=jg}5we|fY`UF>
zB2`?fXz)J(Nm}WE#W`-_AA%-}EmcHm$zn*7;R^PmEZ>bKJW9-Dpi#-=)J_`E;4v~M
zVrL6?kyF4rNd!;Wnc=%o;<uwsP7j?kLg&oTxz&lSXu^#0%|Ro2#*Cg>ooodopQL`1
zdNgPRkDI~ct8X}H#Wm66`=9+%N7oH>-9*=c_%`v=FV%0Ze*KeY)9*c<e$SYmF{fvY
zw*>Po!AJ<YIBz88zXxi*`FZGzkioz4uX_eR{ndRV`Hq=<XEnPMpjiLMH$S}jhr#vS
z?^3@{>8CCm$;$@1VxlWLy3#`2$G1Mbr4PJnpm7t8>uB6#jT$IpqKu9*I~>J^pP}KW
zX!sET@F^3W(%p*3Ax`N*$jO36va4L0#RM!dK?g|ok}#{_3cf~IWZNyDqFTw0aK_m~
z=TxP-__IRceRyH0nKqFpCoSM|t#B6`|ClIFO_Kizt5_+(vjAQ~pt+~tcg*bckI7Dd
zO4&}>2VpnxI$VL8EM3EI!w`ArsgI%AVQ89eQK$9Z+M*8YZf#NVRr{|+1y}9A7IjkZ
itu1O)e@Sgo_w<+79W+cI-l4qa;vju`hkCKGJ^3%qCws5}

diff --git a/draw_tools/draw.py b/draw_tools/draw.py
index e72d0d3..07b3c2c 100644
--- a/draw_tools/draw.py
+++ b/draw_tools/draw.py
@@ -5,21 +5,21 @@ import matplotlib.figure as figure
 import matplotlib.backends.backend_agg as agg
 from Load_process.file_processing import Process_File
 
-def plot_history(history_value, file_name, model_name):
+def plot_history(Epochs, Losses, Accuracys, file_name, model_name):
     File = Process_File()
 
     plt.figure(figsize=(16,4))
     plt.subplot(1,2,1)
-    plt.plot(history_value.history['accuracy'])
-    plt.plot(history_value.history['val_accuracy'])
+    plt.plot(range(1, Epochs + 1), Losses[0])
+    plt.plot(range(1, Epochs + 1), Losses[1])
     plt.ylabel('Accuracy')
     plt.xlabel('epoch')
     plt.legend(['Train','Validation'], loc='upper left')
     plt.title('Model Accuracy')
 
     plt.subplot(1,2,2)
-    plt.plot(history_value.history['loss'])
-    plt.plot(history_value.history['val_loss'])
+    plt.plot(range(1, Epochs + 1), Accuracys[0])
+    plt.plot(range(1, Epochs + 1), Accuracys[1])
     plt.ylabel('loss')
     plt.xlabel('epoch')
     plt.legend(['Train','Validation'], loc='upper left')
diff --git a/experiments/Model_All_Step.py b/experiments/Model_All_Step.py
new file mode 100644
index 0000000..527bb6a
--- /dev/null
+++ b/experiments/Model_All_Step.py
@@ -0,0 +1,131 @@
+from tqdm import tqdm
+from torch.nn import functional
+import torch
+from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score
+from torchmetrics.functional import auroc
+import torch.optim as optim
+
+from all_models_tools.all_model_tools import call_back
+from Model_Loss.Loss import Entropy_Loss
+
+
+class All_Step:
+    def __init__(self, Training_Data_And_Label, Test_Data_And_Label, Validation_Data_And_Label, Model, Epoch, Number_Of_Classes):
+        self.Training_Data_And_Label = Training_Data_And_Label
+        self.Test_Data_And_Label = Test_Data_And_Label
+        self.Validation_Data_And_Label = Validation_Data_And_Label
+
+        self.Model = Model
+        self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+        self.Epoch = Epoch
+        self.Number_Of_Classes = Number_Of_Classes
+
+        pass
+
+    def Training_Step(self, model_name, counter):
+        # 定義優化器，並設定 weight_decay 參數來加入 L2 正則化
+        Optimizer = optim.SGD(self.Model.parameters(), lr=0.045, momentum = 0.9, weight_decay=0.1)
+        model_path, early_stopping, scheduler = call_back(model_name, counter, Optimizer)
+
+        criterion = Entropy_Loss()  # 使用自定義的損失函數
+        train_losses = []
+        val_losses = []
+        train_accuracies = []
+        val_accuracies = []
+
+        for epoch in range(self.Epoch):
+            self.Model.train()
+            running_loss = 0.0
+            all_train_preds = []
+            all_train_labels = []
+
+            epoch_iterator = tqdm(self.Training_Data_And_Label, desc= "Training (Epoch %d)" % epoch)
+
+
+            for inputs, labels in epoch_iterator:
+                # labels = np.reshape(labels, (int(labels.shape[0]), 1))
+                inputs, OneHot_labels = inputs.to(self.device), OneHot_labels.to(self.device)
+                # inputs, labels = inputs.cuda(), labels.cuda()
+
+                Optimizer.zero_grad()
+                outputs = self.Model(inputs)
+                loss = criterion(outputs, OneHot_labels)
+                loss.backward()
+                Optimizer.step()
+                running_loss += loss.item()
+
+                # 收集訓練預測和標籤
+                _, preds = torch.max(outputs, 1)
+                all_train_preds.extend(preds.cpu().numpy())
+                all_train_labels.extend(labels.cpu().numpy())
+
+            Training_Loss = running_loss/len(self.Training_Data_And_Label)
+
+            # all_train_labels = torch.FloatTensor(all_train_labels)
+            # all_train_labels = torch.argmax(all_train_labels, 1)
+            train_accuracy = accuracy_score(all_train_labels, all_train_preds)
+
+            train_losses.append(Training_Loss)
+            train_accuracies.append(train_accuracy)
+            
+            print(f"Epoch [{epoch+1}/{self.epoch}], Loss: {Training_Loss:.4f}, Accuracy: {train_accuracy:0.2f}", end = ' ')
+
+            self.Model.eval()
+            val_loss = 0.0
+            all_val_preds = []
+            all_val_labels = []
+
+            with torch.no_grad():
+                for inputs, labels in self.Validation_Data_And_Label:
+                    inputs, OneHot_labels = inputs.to(self.device), labels.to(self.device)
+
+                    outputs = self.Model(inputs)
+                    loss = criterion(outputs, OneHot_labels)
+                    val_loss += loss.item()
+
+                    # 驗證預測與標籤
+                    _, preds = torch.max(outputs, 1)
+                    all_val_preds.extend(preds.cpu().numpy())
+                    all_val_labels.extend(labels.cpu().numpy())
+
+            # 計算驗證損失與準確率
+            val_loss /= len(list(self.Validation_Data_And_Label))
+            val_accuracy = accuracy_score(all_val_labels, all_val_preds)
+
+            val_losses.append(val_loss)
+            val_accuracies.append(val_accuracy)
+            print(f"Epoch [{epoch+1}/{self.epoch}], Loss: {val_loss:.4f}, Accuracy: {val_accuracy:0.2f}")
+
+            early_stopping(val_loss, self.Model, model_path)
+            if early_stopping.early_stop:
+                print("Early stopping triggered. Training stopped.")
+                break
+
+            # 學習率調整
+            scheduler.step(val_loss)
+
+        return train_losses, val_losses, train_accuracies, val_accuracies
+
+    def Evaluate_Model(self, cnn_model):
+        # 測試模型
+        cnn_model.eval()
+        True_Label, Predict_Label = [], []
+        loss = 0.0
+        with torch.no_grad():
+            for images, labels in self.Test_Data_And_Label:
+                images, OneHot_labels = images.to(self.device), OneHot_labels.to(self.device)
+
+                outputs = cnn_model(images)
+                _, predicted = torch.max(outputs, 1)
+                Predict_Label.extend(predicted.cpu().numpy())
+                True_Label.extend(labels.cpu().numpy())
+
+        loss /= len(self.Test_Data_And_Label)
+
+        accuracy = accuracy_score(True_Label, Predict_Label)
+        precision = precision_score(True_Label, Predict_Label)
+        recall = recall_score(True_Label, Predict_Label)
+        AUC = auroc(True_Label, Predict_Label, task = ["Stomatch_Cancer", "Normal"])
+        f1 = f1_score(True_Label, Predict_Label)
+        return loss, accuracy, precision, recall, AUC, f1, True_Label, Predict_Label
\ No newline at end of file
diff --git a/experiments/__pycache__/__init__.cpython-311.pyc b/experiments/__pycache__/__init__.cpython-311.pyc
index 7c47fb47d6517177c10d034a903442544f29f843..d2d209603266122c16834fcceb26343e40c8038e 100644
GIT binary patch
delta 63
zcmbQtxQdZyIWI340}y2C1gB5rF;@R-1?0F`#T1w1=O!j+#3v`_C8rj}1Qg|`7bWH@
Q=qkKe(*9!c-iiJ$055hGK>z>%

delta 40
ucmZ3*IGK@WIWI340}!xjpGcm_V=VAX&DAQ#FTgn_peR4RC^2_pnhOBT0t=o1

diff --git a/experiments/__pycache__/experiment.cpython-311.pyc b/experiments/__pycache__/experiment.cpython-311.pyc
index b48f212fa4ae511b0e7dea0bb6aa235ea1964a53..df1dd2415f0aa60f1dad3b463e33364a8d9f0dcb 100644
GIT binary patch
literal 9086
zcmbVRYj6`+mTt+C+<I8DvF!MXg$<0&!(PlQIKyiUAtV?KHZR3&uTZzKj2=#_#RMZI
z&V&RJJQHU^z+u=#Oo$`GBPNqs90q2ot@*)L?T>D6S6SOtRJm%9AHT+_B!9MctM=U6
zQcEp6sm-)nx6gf@d+)jDocrB-KPxCO5%4^^?mY3o)dcZZ6iGk26!GmO5SIy>pfz5?
zr*UdDh}L?wzC33hrt`cypWdnW8Jq?j*Lm}Oq?5$7-fQ%koF+^gyk?)pX~A^9x4>t0
zS}{#}ZN5ThA*PMqBHtqCB21gS#Xh^!j%l;E#8>Jp#k9ryjIYdDhUo%txw8WLQ{b(1
zR^e~8vl@RF`)ZsusWJ|yLrd%;XzN*mwvA{B;wJp%$GL<q=pY&jXHm34BN^Q;ua`RR
za-W1q)d9vuQy!n|1Vg#}G}X=q+)OYir!;6KOP@EuQ9Yg@7hq3I%Uc8f?oiMZ@KbFr
zj`h4s1-hxuUk0dFuPYezbbH(`4l;pn$({i4-R-BDKE}^6z<Gy@b3sP=DVNtn<6;!o
z!%!%fr62)iTGGuw@9{Da*ZUau1R}^?EU;_c-Uf7`Tj_)kcLJrg1!!oVp^vzkzEl@%
zyYZ+_Fn)%0LH7}x?J)8Vj_KR$P6-EpG^xn9tAMynFoaV>6HYCyIZHV6Xf1>~IuAlU
zt%J}&>mkgi4ItEf$;`ak$FLqBj3ekyYZT(AhkqXYzcs;3TqeSrFhQYo{BS5I^^Cyx
zW;#MZDP`ukd?gvudzDBmyB#GpmrquC)R%dWy1ZIvpcjU81<sbH2tr*4N|EV2t~gCC
zP|BFjqmdM_MWPg13o)~jmj^M664ODfK#A!gW>sPah$+3I^C4E4&I==+Md=t#LTphw
z7Dg>7{p1?b=>*YRlBT)(GzF=pX*%{yn#%sFrNL@xqhCXtMnIizrd0~fr^1^ygUVP!
z{{z$Ka5zlxSDrdY@{5t#8<(fg-+20`e=(`^>Krqle?B$*n-*2_3di)t_|&BbvzIO;
zKf5p!e}DSkSAC~J(cL}MSI)|e{`O#S=96o)|NQ%8^zG?i-JkkkEIDvt>eAO!fA|1p
zeHoki^483UA50IuEjK^=#&4$M7f|*19eC&UqtC_q7%<q}o^%HKJZ`Ep5b(A*X6_6m
z-@PRF=dS~Az@TPsp9f-U?AFYU5#T-ZhjUYxKAipC+cVK~dtApE@AU2S=)H0_`POI2
zJ74FF(jDStchurYzV{&c?#*QMT=M*v$%l87Z@!z7mm)_XRyeTNfYL+Jrw{+A$RdZW
zFF-S1%J1?qEsm-9wd6+^l0(0qj^72jB;R{yCjQk_^gZ+re*t1nzW+HA94=O$eD{l)
zHy=XAY-~sto}AG%bMEuxy}L8<+cWVfG%Kez<g}ad7pcZ&$;*xBmhB6;dqARiZl`Wt
zgZaxT!LcrnADlVo?xBL7ei(~fSe7w#0%zq^0tb33;0+-^nd$>_udBhmqF!GeNd9^7
z>DWN>?sbSR*sr}I-*NET`?_Vv&jec>@<d)aJA3V;Wc1UV20J(x7lJ~wZ;Va7e{TBS
zKPRuo6)k{9rY{Uky>%g{63v|QxY6X?y`Q{u<>|wJg?ZT3esC&!skQy!Jk4xlxSjx=
ztCxN9a^~@$pnM=G(1|Q7gNA1ZzD&OL>Fl3A%4r)U6y(^Do1?Irq&tzSeK8JIVZ~v*
zIn_GRY@o)_8k3%isoS5<-uM{QN;Xtza^~8gJWH}|pnkj=2Mf*NwiA}hkc(p|+!RO;
zTav6*P-p7?rRm?_Pu?HNsm3zy087Jo@dAyPb{O|NZzj*(pSpK%>fYPaAK#e$@RR!K
zx9=w3`CanrrCbe7F<?$H?CDNe*(i<)aumx1LtYLIYIt_=ELcYJ(j8Dq=!iU)=|2ul
zeRTEd*hq5pT=LGFGY`g~!DMuF>hp_J*P>6yZY$lC$n+}RzbZ{RI!=204ldxp3hme%
zV0|vH15k(0<?eB~U4A#i&NhSd^%VY$AZ7=EnngZVDm?BDxKFNGM+KQaq}R<7A?e&9
z+9l<=`$Bsg$|O_9Ik9NDlB`(}ByhZ!MXRi2-NkSiqfl6rr3w_^=Vx93__NcGB#P!$
z&NC`5iA5{1R4C)8jIyz4iIYqj7bDqn%wDo9xFpMZB|U>IUMkErYsuhsoo3h|20tp>
zm5SyWqohOTEE#00lk|8=kPNatvE`_17Eh3JopO1+uHyi-bdoN}c)L-_fH%mZ6;Lt+
zvCRh20Fih7{^vmKXz5B#2_;WS7XY>stjp(E?U=cKZswDRU72wPv*^aY(-KKhU>qDp
z_1m)wHKtSTz(4q35DXA=EkvawsvEQ=7jG06ZxR=8it0qN@=1APoV)e<L{r;%Q=8DV
zPi)%9?>{K+|EG!lXU6xR5%x#K{Sl%3HL?6PsBC@W__^TNB09F9%GFOwD&9MM@i1hV
zCM%XA=82j%lv6dibUFGiNt9He+=QbJS4fmT15tCr(JV)o;P2u|dnqIv6Qxxs_ep8x
zMepr{x11BJcZ{#zA*_C0T>U&>w?DQaw&C$kq4a=QdH|K1BrQ?hw=g(Dg%J_j!5HxY
zAT+>;*%(hj5U5uAff5QjRO5^;;1khB^btcgYV1{TrW&h605kQXsl`uhL3|k2hYe~}
zs3HDeyh%>MZGiDKVKSWG4+hZ>h*I5W;HRKCga&D?y1j@oyO+HR)>Ne`*wT%VUsf4P
z$x)S8(5EU@L7|vSR+$Vwh4N{E5itd)DoH>9#@Et$BOvpeirh8oSzWjup&FagZ(8?v
z@yhCbuQD!L|D(LJW-K!zway?70|_fA9B5_sfL0JVSC`?Sl9__TGx_uP3cbqcp{|0}
zfkyrTKO>;GHx=E5mKA&*CSnR=PnSkt^*ACXt|^lXoU=yMn>C_dWkg{UlqyTt0UG&G
z{EUE3--N&XM9eD9H~g69ht08ybiM2^aIZ|q!+C0Z=z>v4ERlkUHDZJMRq1ls-*C?U
zqYEq|QWz<UEQ%CI?2(d4Y2+D@6P$(2f5N3<$Wyhik~P?CmYx8Y#q)W9jueKA!i&Pi
zVSBg)?YJ_tI0AjSsaR;PF9=)01t+wTvamI5gKs&vHNzdaWaM8KE~oRk?OC~bYF_=2
zrD_-a+fBvCVT)9B<_CPA3HLnH*Ib*<8e=y)2q5={o`Y89y$`}mfEHkTlR+9X+^t`;
zeqBTI<9mPrr#}4E?4|er_Tb90ZpZXzzcsOFfD)O8kWzl9%tbOCQcg@lYoUzEWXj%R
zr5JDg2Y7yfwFy>sgiITjut>LLtkdB-#ZU(V0WQd{K#4jIpdMR;NJ|RIV9YG(`&f^k
zW093WA?TE3!hjI41LKuUnITGr8Jd!F$qqnKPETJ@%13w#D9Mr@A&!|-qr?$wW~3<2
zBIm^-O=W9Q0Kku68z7KuIryAKX3H)`K|RVZP})F1oLz#_tZFzd6)%93rK%spzARE9
zDGxRV4Hax7YV28*V8*gd4LIK&2%!@#NsqfA>CpJtji{U+5EaWwd2kpe8F!^mrQkTu
z@0WKOfWn}+k|o725DYRwEMLki(<xZ!a=SyU%k6<&1ZAnLxDJ(+$W+Q{$=C-2@}L7d
z$p91_{7HG+54K9$?lpjm*$|ULv<2-fLwjJmBLiH+o>b%8<)%)<ks$I`2na6xuxro|
zZ5arRb*H~F8|md_z5q3XqX1_Hh(zu3>rE52+s13RJ+l9$QmAbcYuf~}CaRBeL(XKO
z{i5+b^F?!1pCB!R)(Nt1oUDuOh_4XH^&+{RC)Yo*E*kWlKRI|Zn)ie(;H`W3_748g
z%R>9hg0)K^kBQ_lo;;Sc*TsGj*9!KPqJ3r5IA<W1uekn?;<9JQmOf|{TRM2M7Gb-g
zbr-iL)~x?(vAE{N$2<OGuej$VPc}l1ZP+fBH=;h+CP>FP>4=rY&7)3%+$NIScye2I
zj{SOhe8;VQV@CyYmq_m7$zAH)n$e|V^RCBL0(nRz5Ao!o$<|$e**ekc9B*|BtzBX(
zbbA{@p5?>q#LAWNPI2|Nq`hjm^C~q_{oHu<b3*lIv3m0u_uzHDy;E$b`R-n^`;1@@
zi}o-Y#iYG#!d^dauNUkqMEi=U@rmu3VVzjk5ML**d~U2$+}zF|c$q)u5|5qWJwDO%
z`ZokoxLx}l4x_s7Erij^TVHtG${#rV?=K0~BLaC;B#-js(aD{=Cw3kl-+7on(j^{&
z+I_;#SHzvK2;?uKmZ4t0ws~}KqP=sX-96s!=4qeM?ibtr0tt9K+B0m7ZHTWO)sDXS
zr~qJbN{|)tV`DoW?0bAzAP<V<L7qGa^BAQktWD$ArcvE!&m-gG4SdHT{)m(Ba`TKw
zWKIj#e$m=5kY_~l3{Re!T(mf5;)_;B4T;j_aVuZ4F=~23+J`zXI%B2riqXRlgO4_T
z9exlN*1z<)Q*3hy<Z+Qa&XdQ{%or!g#pC4S;nVS-jjsD@lR$13$;~{udD2!qbOchD
zkJgQ{yloRtZeo!ac4y;rg!a(?+kb=LGQnl#<zNpDr{H%uUsUKAsy**f*QGTlVKoM?
zW9Jl%1U!^=Ev!+mf)Pzv6H~BBHgi}Tm6PT9bzn8qhL%Hl`QQb1`m+Nc!cISRur%_9
zavp~Fw=9BO4J3<@0lN|f@(xqpZc^AEN?N~PvY;D)b{7jy2-XcP>`GUmnp)YFb^skj
z@^Cn?FJYetTJ1k_oPH41*bBh`k+d%r?8`;_@~AOsE*xx`FxQToYX$RC(Y!SF%DB0n
zH`hNYtGH^58-=oFv8*||d(vKgdCzb6#O&8A1^a5zzB+15*vfdaOrBjus=y!7cy<o}
zp`KlY_|+IKqR&3t2Na<X&qwk4%*=zdH6gUd{pblzLmrEE4lJ5ZNyjoSZ-W8LO)_V^
zI4lP$Hy)H1KD`HSVP2IC@_|qgv5<YZEa%-2@2|=B(+ez&5hxT~4Z#5My@6O%8(SHF
zO(@zT7Hx?do{)|4{;>^@)(hlrk=)IbyC-cWgJ6*rvM8192vkRdz@m(UiKs<60$DfV
zFFyd%vrHl*uIvqHm{cn-t%+d+N%Kg5a}9&m{y@1rI<FrlJBRBJm4-W^pHLb$hCiXS
z;U|^WFIZZ4H9wqxLUT)>qoC%nX<^Vy8^UIFS%Avu$|66s31s<qAN=(j&a^Z(cSrj3
z<?pgt8rU`DWA$W@0?odRf-V#|A&@lP>@oN@XjmVLrVx>Q0tt$b+@GZF0~m!|%JXpk
z#IP@+G($H!H{_%O3~pOH4pI0b(IUGT+!ErSX->*>h1~2bxGpx6l+760@ad_fL-plm
zx=}Mhq`<QEWZTe-klJ6rfDJ8phWZwi4*n+u;J9p+=U*Fy<Eo^+^x~dm-SSxfwXN5;
zCaRZxOkJVkokDf9SlyhcT{%(PJPyn1rm^7NtwL?9SlgPYsh_B68n0;@HH;PCwFotv
z#hT5DB`YSDtQlXjX0(2+<8Gs{WV^Uzd!nLlqGH8(#ftcvc<{zXp<=CAu{Ke)Y@%w_
zc-5+S$LQJ{M}?}5V%5eu8(FkyjzB>nQC&Cu>|C+2WZ5?a1mB?mFq54y7EF*8<7CC~
zmN+bvDF6mos<1rmJ@+<{=(G>rtx%L?q7b|dc@b@)s8>QRubji8^IX=A0_5yj<cU}$
z${Pgs9I}5xcX=!dis2u0K`@~F=CsZFf;mD7dNd9CmN`NR4rxezIgBSA*bVy1IYJ2v
zi%`b=KyDqi*-$Gv&}}(h>4FqKA8k|;<T#5Qm9z|Bq@yeJ^yUd43OAzU?S8slzKD>|
zi)}RP`elk4BJsvT_U+a9-aN&kG4~Q(KB}&zGW)E?%$BLMhuMj?9>^x-gYGI7vu;r3
zbmw4qDqB$zzAK~L=>EPD)AHGMaaI@c-Mic6!}~GG7Ch-?T&y2T!v%voh!-ZQJX`rw
zw~+F^1qKFE^=^C_pAti(98)$y^5)XohOgwdr|#(SrcN^SflOUN7OnP@-XHS8NwU`8
zhmnRv;-l?Gzn|@fY_<gjSYKX8)LIm*f#AQ%pUWD!g&l$@LTAC3A)L$8Xfz381FwD)
z#9m(gCWv-^p*KNn<<)P3c!5{H31U63eiK9;uYMCmJ+FQf#429>F4WIWJfV6Q;#|xV
ssW(9^;?r+}ungc|f+*qDZ-Oun;NP6CRjUC5SvdR#r+@bYxmfuB2UO_(egFUf

literal 16356
zcmcJ0ZEzdMk>~;hzy-M^0D+=NiXR}vFGv(WMalYPc%&`Kp(w?aEgO=8g0M>pB0lH^
zAWJlOr`U=`JIskrWSe%Bgt{mS6)6stI5CqbIo;i>SE~2&V;8kmK<$;PRCR_V)>XY@
z%1+f)-k;Y!3oHN<bevS)a@jjO-90ng(=*fEGx$+SNf8AX{eNdUZ#PB#2fm~)eJb(k
zYZ{7rk76lS<D>i<r^c^!YW)Sy0>93w^Xr{@4dS$bGdK<YLT90$cG7;M)95d977<*5
zuh`Ex8GnhhguLr~rGAss<TpFb<X!Kx_}4kt`K?YXc{ljV{Og_TNxINi?%&|t;ID92
zkayZw>92BDk+jjb(O>PXCg~!mzs6bPuXWb?>zs9do73j6ch>vuPCLOD1AdcplfS{)
z;BRy``kS0h{$^)0!7;uTzr*S9Z+33>Z*gw%w>n$>Tb)}8uEf{oZ+Etnbg8ey-|6fm
zX%l=?m$Qqc&Ax8`Hs>~yw)lGd+nw7<dYx~F^C^(cI^Ry`PV(I4+(n+doqNc0uX8VX
z?sM+bQu`^^dXZwwxUL%o6m<*!mCM=7t{bKt>zDA=p_hwa*uTGXr^_E?IiGA9-~u6T
zhzoahbf&UOpW{LSF5J_RdapXndq=$ix6iejwyOWMJ21+*2KeB3FcNn42f2|EugA*;
z!cbFpAIrLUF63n+z<tUW^o+UO0oE0MCFt_G&vSgp<qf!A^l;;0Z!nOmZrd76Bi^&z
z+FXsu<qn6TVk}CjYjxrEsrEj{jYfQK-aFwY<XO#@tFWX#PuP1_!DSWV(8^WbfG6np
z21Z>6IDVW9xyIeR+s}pJTjUxQcPJPb<IW!mp36c5GZQ);85!|$F4i4(%j-hO8|GXi
z-g6vF@_@$Zar;14ZqJy!!3`}5f^7WW36#}HBmm6{1{@l>`Z><cy1afj$Q~tngb#YS
z5Qr0Vu%>ZeFzh<*4TXdJd6{8(_baZ`oIC7yj{|q<h!+~6aC;(Q$f-Q*_IaUID7)}!
z&IMxRc%?}O*^2Gp(-Om$40h(r?2^~#p&U$QpYpmxIi<*@!L(u#jg)&(^7+)|G|#Qw
z^`K|vDUEgq&MP{q=!skjnhdIqn!E~?(YlqW>Krv+iQS-%hD-@KlAsypGZ7a0g9w*9
z2eOxGzk3XN+sn(PBo%Ul4)eiaShf(_=O77f8{-Cmq=aDmhhHEiK&R;a91p#zk?H+B
z^mc#WAxM{a1KzOfEa<NLln;?+PcSeN36TNh4)fk~$ax5xG~`Lm0OX-fy?lBa5|>~`
z<|wCzrJUL+YDB|oF4E2dR(p|h>R1!2yQp#MSv|ZNSOdU9wveMov@Cs5=QOfL$SGor
z02Z^w02!76Si+V7EM-dpnx?388mAd%9kX1_og3$PuOCz@<bkC?y$T_Plhminko%72
zJt~}qy#a_@D(PGnM$M#hX*5(gz2pE%E-p&B)afv6T&o12QcDPzd=3@1oLpf`I+dqJ
zI*nsBP_kMptA#hRY$Bd5fVX0mQU`C0`lg3BwH>Sh-bz(mA-tK?HyYl|>9;5~m({Lx
zE=5JDGiobtfJASpQU&gHX}WN8Iz>3tI)FQ?hD=G2uYL`r)pBltEN`jugc{1y<+4Ul
zjH1X7p-YNvSy%G4pC`Zm!{p`HAHB1<D+hx_n{63Ia>1O)g?8DJ-~af@wdqH<uK`CM
z24#xgWmBGy-@N(BZ+_Wod-8*8OW*yc<n-5*Gq+k!jknsyN8tagBToT-GLMf3&+W3Q
z$-Kq%pZ0pjxByVL+SJLv)ixC5{cay*aqf_h3+1cAeJVhR;A5uVSi1JT$FKeR@jDC2
z#l^=TTulD<PV)83Pkw&s(Yx>GX<aDn4o5%=kM8_B`QZnTrrv&X<7<zv%mbssvh<xd
z9?kxA+zrET>9vcHM3PHCPJVm`NOmaIsgxj>Z~PD{ht((f<|Rm^2wIiw<kx@s$#32T
z3>2eCos^qhS56CaUSORgOJu%IC1!{#0Ba~n;PSi4nQ#8r-Rb1pZ#|j+2%3xc>`&pN
zlHdMu^1?SB&0kA??;9=2%Rhbc`fmZTG&l3){dXSEyrCf4m9Kj;|C1;4Q_wUeH=oo;
zxc~=iZ;*$THtad=3c(5sg8b>_PZlpcdh=JHQ6S4FA6`ga{~)>W_M;zPNnZN(C%=0m
z=c`gmIUe+!-er3-|Lw;gf3&71hIu#G5!rR67~X#kDoWKt>V;-2-40#Q`S=$X^HjM;
z;}^dJO-){X_c3(RJBv$~E+-ehx&O%XPv&PIz46PXTUVfD5G3Tc+8!yicYgtdipJ!r
zM$MAx^ins`^&IC9p5<(4pN#UsNPxA0`C;QZx6elG3EL~)@M)W((n`)m=W3qdh%GZ>
zR<kUzFjaU3m}Zx3km7OaTG9a+a;%s2p$I6nOuN8#gq6_6Zvcc`nq|NQPo3e*G1Caf
zjbi5TR(v-rCeOHHEP)Znw_tWrMh9h6)+B>UDI3XhPLgH$1TX73)BxGMvWw+H56B11
zDA}Ocx^n3#7gj72qVHtoJ{$B!h}DHP@D6N_ZWLB~u;Fwe&Nm_(T`rH$4W^wdgkrE=
z_)}}UKM-!`+J}zxw}agW)c*ENt3z4ow(;|P4=_!XW#QY>f~x{d=zjzF?+er)?Ub?f
z&BE!zI9(B=D@3|VqN@bDDp|WjtlcTq?wry|^u`C3&0?iPs&q{0rb{2#_K3E9l5HQN
zY96ewdi&UwV}K}nSk(XzQ=)DdG1ZC6K_pK!G~;W1VtpkNC2V%Wm)KANuS}wR13Zc!
z+8W}v?wGB6Vc?^~qHRF34Jf6TRnH#1dSZUt?cIyIyQceP(!rC0%OknMVp&8givX`N
zK{HdjPvLVY52{M~!nKuH84RdXu8nafO|!yUn1ogMjJh_?r41*Q8WXwp1=t=cJw>fq
zu0E_<4zTiTqWY+A0@lt61C*Du9&2!|R!cmieos!Wy0&`IqR67jSTQ;DP>XWq<f<(q
zrDV~#R7|#&6V>Flbby-Co?Ms&+hk6)Ez<SiGgnTNg{*dx4y%>`$V#<+=2W{Un?FkD
z*2gzQ3+Ggu=<_I{Pt-3!J1e&wz}95ASskmt0sVYS?dwS+)T7!y8A>S05H+%eVOu8e
zjB0Josn$;h3mo)qV{YwhaN{Jk>SBw&Aa|BjGlj6lE4fE?x0&1k0nxVgX`VTpI#*pF
zpBxRHECTu9%)9E!?eP`8Q1ZDtUpQB>syN7q7R^=W&DTc3dOm~m7`bkj=6)C8Hl}fq
zuFgtl>G>H|OGdQe-Dw(P`_`u4fV(c8&efCKrd&IBvKU5!Elr#KjDp2}CVz=AKV`x-
z&>Mww_1T>1H1KDQ@-yoA&u!5O$e&~;OD0PvO_S!Ri8YTAtqvKYrRg+Zz?vs50Sjx1
znxdr>z`?GIn(zs4P~8NKN_?BNz}MK*^=Ch$Otd6g3f$(X<pxOemMT%yrnybowCVJp
z;b}<IWk0idvU?$0mN&rDx}|zspmAT$ln^aut>Hsi@7X<>qtW=?Ii;@b73xfLx}5xL
zLQel+Z-WGt3(^_=fj+#Y_9IH`oF;>1%O+rU^JrZv{fu_{XW26uNEfY^RaX?e%rj^}
zkSkXwa>p(P7cn3>6hU=IEBErdVJ;K~gXqcBrHmn^_y#SVJ=+{9N9+C9mfrpLf8G6{
zam4ocN8c)%aJRL!!%YGg7&bZur6^&%_M^dZucv*r=|}8C#V=Cq!}bhQi><4pt9xr_
z$JWjrHpkYkw$nWkoa?cPmwx@|_FGGrzF9O{6#v2a-l8A%F5$8+S9l<VHVTm>h?}`Q
zd@vM(Km$Z0&&#yi<B9NY&w1H6&T}3whDabbq9eAP0dm0Akqi2s?{^f)Iv+5~Mm5+b
z7lOkAWriXU)~1SziWDg+Mte#wJ38Q6Lo!8BQBlU+cj#mTF%v`lFgDth5wa2sBvC&!
zszb!s9R(t|KI=^-gkD)s?8_xkS<10l)}aH;?|>Hw^`&BeesD2l{W$LpgaHpdAr~P8
zi}Mr&dW$ku%H~Wuh1K&I#+6MFOA3Ygh$q#HvNjl!b>r^vX=p!2v_dlD3qmaF70x?)
zIvnD0y+cFSKMoN^Sw99&RfvBMGmS^mGM5d{5n1zxA)kK=k!B2lLXZ)H#~uzNLvmp#
z;`h5DR#`COh31d=g0Jv<kbW<MeF$)s<#!?I1t43Y{2<Rl_zZ(yikUCd9DG^?G(s*T
zO39t&coY?e6(c^V2ANo}$*WV`NO1lrQV>}wWIQgVasfz!@5dDAn9l?8Vu;m30FTF@
zYuF8;xiG&I(2jNd08*JVMz#{4;E!Q;vEmpgy`Tr5oe#?eT!57|UT{zX>9Ds^aY8Uw
zXgr*b`9cp8$C1Q~3sd1-{s<Jy;~d8iA{at|vXwO>A?y%zOmi>(8AO{hjt@x98<^Ob
zX>q7+1rlW(;$VnfpdOgZt{C4gzEV7;2P1pBG)`}d(VIlNQKB0Ky758jx@qs5W7A_(
z1rO*Fp>$9<GAz7sTs(4IEIlF8Cnfr%K%ZPTP)(b!e^qLHYO&#No3v|Kpuq!}HqLCj
z(wpe)`KVUveD?mp?}wzpF@bIajA^z^s%%SI%7qQhqNPQ$v@BDd#@at1nCeSZRL3h?
zV->ApMVnO7Hg)h}#j8^XADArDuU!87xBh<CE1Fs)6R>DYS`y~UxVb52ZkjtUnmZ+P
zr(o`UU|uhjH;LwE$=ocMo0Apo3;KnZ#fqI$#ZD-USs!O=VoXh9eS4z15!zn4=|M%!
z)slEcdkjdA3l;5RMW0mBm&(PKEVKyNoZV8z?o_TLR^eD!FH|_hiXBqLj%A~MefhGH
zDuItEDp~g#f@K0KsH$4{*mz}ctg`o>My&ihsq*ipo=v2iAu@FmQztNW4;YKEE=$1c
zPKeA&i8(1SCllp$@A<#wPnfITGkwc6r=NTEUYXSAoH2>!<C6J!qO#^{RidU|s_9B>
zYQEmPFd%I^DA=EUP~AAkN=>^K2ks7v)d!{OgUiL#hK4^-g&RsgLol^}c^y?^i8Bo`
zreW@-1)bC}aF6{TK9L!gm|=k#PW0@H_w0}L?7y%3y-^rGCJmnwdpuH)M`XID24-4j
z56+#P4}lEL<+GNzUY&Y%);r(%L0~qJC|f_HnbAC4Z=W|TREjOT7SG?`F7|#!IC@g-
zJt;PSRV;G>nlZ<j#ux+5oIkMOy;pwUdfzQFM<nKmz#K_1=F5X`4bGO$o|~^+cxJKx
zo<(H(C8l3s`T-Ybw#JyP^X$TbMeqG`;VVal7mf)pos?eUgwZi+^i`3GN=#H>qDiJ=
zcEeorPx=>hA2DL<Ua57jXy3aSmU@Rp=BUIR6_}$?-t@sZV~;WRxoy{X&U+Wj?$+IJ
z_<gI$9Fv%10&@%~rUs`6XZu8^1`}@%!ibn+<E2|-rCWs7J&Ww!pjbL2l@5vYVTnF0
zq;Gyd4$wk(APCDSpX0!y5rHe+r+)(Q9-1idRIdDC@n`|7$zA?KJ;0^9Hgj0oste2q
zZ7vsagH=OiPBmt-S);mK(_$Fg_XTQVwYjGHq&})=3o=$dyy-G;BU&(5^h2<sr-G$R
zSOv_U7#Q)6obQB%E0t8%vrdPB-;N}ftd$O?a}4>A1;~_{(aFTDETtqslbLa$5M^Qy
z2g6`oD8VCT>rP1r1Ka`Fz)3Bz<e@}%iWCk)YgN+erjzwPgx0b$Me}58u2b7Hh^sZ(
z(}VTpD9-)^uvSiN$)2kJ4Dlos9(o<X1u9{(N+x^Uv@K@Zw&1*1C7K2$)8I0tH=fYI
zj6mj`eX+`YLhoVWD;}|ul`2`7cdQ$(?2B7FV%Cm@D$%+}vhER>JrC_ol6^<qzCUK)
zFC2JLaQcKmP_&Or_Hme*64s`<S0!t=&~rdI=#+Yn3)bTTbDSu4etkySf<A-ZMGHm?
zh_*qFa;>4+I##n%b3ii+hKSL`9|T^9R*C7E$le(dHd*T&dZjN7FuY*fR)*ykw`d_?
z&!eQ|VnVO%!iG@QN$)6~Q>K)K`?47wn8@0W2({&-_&SslLQ@^OVMERA`S}*1ynCwn
zp~*Tu5jQzvCdd4ig_jrWMbjS1v`3)#D6&EeFMa7CLB>8V^tr(+7~?ta%Ly0~lbT6w
zIBRsHY;Ug|)dkT4lv$(+id1(FmM*-HF>eyYR(27TfIBftM!U)NAH5KiG<XB!5ikP_
zVYeR>An!sbh;Iep&?z!5R!kW}DH{|*9^I=Deh&(k$dz})TdobV!u)&-d9dUQRI>J|
zMXOl5SE}7B(i^9WXI@B_RnF~RY!b@)#IioAtZ&Nr&{QGN6-0`px*WB`5l2M%2ok7v
zbMRb!!StneK(2r%b=muiXH+pqK`B=Wc`e#34?Yi<KCDK=z!%j;^`j8C`X|k8-HOT5
zki9z@HQ)fsr2r3451nj?V2msI*6Zj!ef;sGkKcIw{hLeg|DfgZ8+Vd#eFydm9Q=zw
zfnJdt%zH<nGLSHjzlZ>b6pt>@5{`)_6e18qKpQ5h(2%20E(F(#^?Jhm2|&r(z_@Gx
z%h>Ne$Df1@ehUH&EkV%(m_{`qR<o@423YQ#42~GF%nj#>1uh#Ddp?9(g{lIVqAdZ)
zn5fI12J<)o-vXjg7{CSUp`|KrvBxZAtrjg?CCk<+!vl-;O6BbK`Fg?9Dq314OKUoB
z+tuEMPQlV6T6!c)59Zd-HO)5(JBL!Y<*;ZuELjc%iY}XZVb(dnUZmS4x?M=$%J|I=
zgEG((qI4HPsdC3}M$NL-YGt^S)vUEz=PD<wU4yQGEx<s|7o^txMbz5Ah}!TMQG+#*
zr`@`%h0(%M&Fxje+U!kz4AACUGB6wpVK~qcbk|@%Y-#!;Y%C_<{Sf9n+`%N%-s4})
zJo?_%PZn<^7cM0~eErGYMHn^7sRh{ldvtB;lf|D*7+&4g)-^IYQK&o{`A#VAW+72$
z#VNySTJgt-d-f0bik{g`4gUjp$c$oKC@Mb$M@F>ctZc*u{uw+)#D|c`FoGMxVZIq&
z97W2o$>Z|L1@4H4$0~4AhM^Qv#uX`U1d9v7-U3BN5CO_C;vsLMCC@O*>j{4V)S-U`
z05fn|`IW)srlz@xYrWTd6E%(BcYWZRKPuLAp-W-k61R87>|F~x7ejY?MSH(w?@!dV
z#OvB)b?pm=Me7|#tlKNq?M>8gj@Ng_>N^)&7KiV&iS>O_eP5z#Q@m<(tZMUo=X~hq
z4za3Rs_F*ED_-3it8SeiUg*C0l32Y%s@}0|qAlPwAuv-ln`WO{wi?$r{(%DU8G<R}
zAIm6X`AlHWE7Bbj-65oJWpv`SlD=?clF|7;A%T`P1me;-SRhh?xJ(Y>)x{yIP2n>+
zh*xdjXh8~}$w55UM=sd<Q38FL9K`EXygr4`<RBiWVRA(cDZHA4LA{%rA%_9%WIQT|
z5iJH&Y8<Qt-_%%`T!gPMiDDeAkPt>ZvK|v7pw$CVP2Nz$hloJETRlg=47n55mC6P2
zmU$${860dO*rzL*E}1P8=^9MHyFgb;bmi=e5K>T*BoDVB!H=h#VILBb$Y+A6z;=<Y
z#{|AJ)68rE+5?cl_oC^d8TXqeeB!I_bNYNAnUTMMH?_jfBAX>@pU$bh58tL<upC4F
zEfq(2)P7!rXN4z4c+|dMgGZaE9#tRK;4$Rk!F8QnYw)bt{80kP{Lg>|On{<Lm?hc<
zX0^dRK8~@i63oleG<;2I8iQ$7B?<G$Fhue=hyely;<5a%py1WYH4eGCR(HM--yP)r
zM*u|ArcH6WHb&RZhDEwTq8kLd0TctYDNfhN==wQ_NViIKt3bCxwn}_fq?;tVNuZku
zab=vg#c10cY>RJ}=*<GXIh*)p&_~$U7GP@|ieTdO#u&YEwqK-e5^WRGx1x!vMnL;e
z=4AF!G%;rbS*6SNrgQskQnMz9s%a}Yv}<yxdbWb2U`-BH%T{pc*5puiYz2pYO%8Rg
zUcq5dd`2{oqlOW~Fr{cB&0hnNV%V1De*-E0w+McR;GYrPL+~#EQnskP0fMms)(tUa
zmOJZJ4!@FFvpNqVqs*0guqoZ%m<Oe*dRZRih+Lj07skLU^_NRjuiph<I8m8B%jTfS
z=y?~u6C$_vhU-o9Ww+}@`%cNeQ=sb}G&`=JooA)CFNw{4Qga^+hzAW#*Pnr-6}@7^
zZmD56DDs1*mg^()N2RU%#HL=UsaK#kJ*ca{dUlSz9u(`gOLf}?y81!=rt9VNu$e5@
z_ek|U0$uyi)FiYF2+z0#cXXMe{y{UWg~#*SF?_%-BtYYsRv8JZjzT+ei-(Ma|Au)r
z&pKQ|HpGBPxyU-586lbp1)s^uU5C?nb<QUIYW=XzEBu+9+;upOSH}q9SM^`z&*V^Z
zs;;NojL#eJe@o4nEC^>EFF2!aUxuwJ2?Y^3_^Ql00^R_cbgQoHRScqk+Dzmu*A8Y2
z*5p^E$Li4ThEnu@MJcL`)lv+9MJZ}esHGJC6{V>CrItef6{V=Xs+MA0vy`H%aJ&V)
zlj2;dL*O9mpajc6%oOC5Icfqcy$-ySl9YEc_4vd0lE3<S^1GL^{Sr77skkP}=HqJb
zBvJ}$p*SZ0dIz4}Ba_6LkD!*N92OOUjtlYpS9tmScM*Zc8h;ajTnfe@+3r$276-%s
z9C5!u@Jj?AAy`E4F@pbv0Ck+djR4&l{wD~CFQbIu_y}g~1OUH)STzCj3xHS=faB4j
zpJ>QG;iX%sY{>l^a$g2z2LB&o@rR}l2{l_qW-F#3I4>}@5>qSKyI^ZVNn^zR@bqD!
zu3co>F-53qB&J5F?-ZF%fx$FVJv;rZP_<cPHe(7iAighDL+rm5QxEGK<MrLK`tH<h
z$kZk)s;)9w{?gWX&9+$0wuPe~ofHOM5RSbl4ZI}QIHekA!rmRX?~2)XEpC%`AA;R8
zardzB{HSo+D?NWkw2w*lu|!=*ylzLVZU;I3wn=>IkZ{;7obpJAS+S0j>bOLGSG;~_
ztbXTWgS6|*_dVjSuL#4OFfu9)pBC%AQoT1(wIyEF6|3r6*!Iz`dj(>5zwq=i;l-Dv
zr=4QeajEKfqIzq*dRwd-n|R`0zu2>1ICxw*aZ)<?Rk7M7Rl9P1uiDve@VzQ+;Coft
z!1t=OO&tW^tE38~G=ETJwoA-*f!Ut0RtrqEGU=<+1N0IOPaNo<{vQB1>1SN8HH=XB
zkwWgk;k7Jm416KZf>oR_<h~H+icka82?oa(;w)TMCq1m3yf6;QdgUiC4Edo_j|&2y
zXSd1pbE)5}$X3kgQGUkM?dm3Yz5^sBn+Zo}mn+1LBT<)Job`JZxh$Kp%hg3VAxMk6
zA^ar(kRHG^4nXBZX)3<f2N@Gp*@G<ixaz+n!({+qSQne5;@Ws|N30k&2o{ft#ZODc
zPcKs%qdj4*pV_wp?-Yv<NW}*b??_l%1Z&IM1WgHRqhM`ZL2yzm{<2j3Wh`NH!s-yL
zj^zSKl6Rt74h{b`Xqv-<=YsKohsz}wx$r1T1b*q{k{OrlW%v!6f+1&O`3peL<M;6I
zAaEnVSTm1qDvx0j9^F}f908fwa2n!KOma!;P_BoAJpmU~{xsx<GyuM&L|NL96ag+4
zA>6rxL$>L^W$h+?$ugA(JgeEGKcZQtvS3tW(4T;;RnT0Z-@8nypva)FM5+`p8}uFX
z9A<Q^1ZGCxxlE~`x)2+qf^`OcJ#wUgnbxD>mIceo^)<_s3OZ;#^uy-?(m_NRjxWjW
z%H9BIsdCN)&g8(!12|7X&N-|`@;FEEI9c$x6!U0VDLW2mG4U8|<<TbMixFVFRc^#{
z;a>PXoATqn^uY#lTs)J_qt(fmA_M%k1AZ-~>=k8BL8SR`&k)y8-b`q57l_b@T;X8Q
z7iv>8ly~Kr$~vTNBtO!_pXa5I=#cHBHq7n|ut$`0JIc{?6ZwgrQX(eF5fl<Rm)rha
z^IQ-xbY+gZmt);__=#Bh7kzCnsL7+Jxu@|=I))s{!Fe8xUy4U9<*N{EM1UtHdE7eS
z@w_5mhrouQ9)TTzTs?@#fKuPpM$X7*M8|JJx&{P|2$~R}2_RdPjWV)Zt8AJn+jW-o
zFWlNl!sGUnv+(?8WHN=ue4LvP00TLbMRpZrh8#9hgcBkoS2m2JKNnIgkWpl%VGEYS
z13<Dq5b?ut_-ojdmk^vnfJ5{qrD*6gkp2_B2c!*i#y<W(;T1=8=#Q{$!E~e1ENdx^
z2fdiTC4@2uby()#4koBEA>W;#4hi}01a(-*cPFSJA@@#D+l730f;uSV-U;e}kb9%_
zoCWx$|DPa~|0PoX-wr0Ii15Yk1hq-Xy%SWEkb5VnULp5RP_07louF!k+&e*8h4h`E
xm<!}DL76U)zXY{TNZ$#n_yYM$P{s@7Z&~+jfd;1M{BUOL9|-2NHSlCu{6C?rn|uHO

diff --git a/experiments/experiment.py b/experiments/experiment.py
index 77d3145..cc40169 100644
--- a/experiments/experiment.py
+++ b/experiments/experiment.py
@@ -1,38 +1,49 @@
 from all_models_tools.all_model_tools import call_back
 from Read_and_process_image.ReadAndProcess import Read_image_and_Process_image
 from draw_tools.draw import plot_history, Confusion_Matrix_of_Two_Classification
-from keras import regularizers
 from Load_process.Load_Indepentend import Load_Indepentend_Data
 from _validation.ValidationTheEnterData import validation_the_enter_data
-from keras.layers import GlobalAveragePooling2D, Dense, Dropout
-from keras.applications import Xception
 from Load_process.file_processing import Process_File
 from merge_class.merge import merge
 from draw_tools.Grad_cam import Grad_CAM
 from sklearn.metrics import confusion_matrix
-from keras.models import Model
-from keras.optimizers import SGD
+from experiments.pytorch_Model import ModifiedXception
 from Image_Process.Image_Generator import Image_generator
+from Model_All_Step import All_Step
 import pandas as pd
-import keras
 import numpy as np
+import torch
+import torch.nn as nn
 import time
 
 class experiments():
-    def __init__(self, tools, status):
-        '''            
-            parmeter:
-                * validation_obj : 驗證物件
-                * cut_image : 切割影像物件
-                * image_processing : 讀檔與讀檔資料處理物件
-                * merge : 合併物件
-                * model_name: 取名，告訴我我是用哪個模型(可能是預處理模型/自己設計的模型)
-                * generator_batch_size: 每一批次要讀多少檔出來
-                * epoch: 訓練次數
-                * train_batch_size: 訓練時要多少批次的影像為1組
-                * generator_batch_size: 減少圖片數量對GPU記憶體的用量, 減少張數用的
-                * experiment_name : 本次實驗名稱
-                * convolution_name : Pre-train model 的最後一層Convolotion的名稱
+    def __init__(self, tools, Number_Of_Classes, status):
+        '''   
+            # 實驗物件
+
+            ## 說明:
+            * 用於開始訓練pytorch的物件，裡面分為數個方法，負責處理實驗過程的種種
+
+            ## parmeter:
+            * Topic_Tool: 讀取訓練、驗證、測試的資料集與Label等等的內容
+            * cut_image: 呼叫切割影像物件
+            * merge: 合併的物件
+            * model_name: 模型名稱，告訴我我是用哪個模型(可能是預處理模型/自己設計的模型)
+            * experiment_name: 實驗名稱
+            * epoch: 訓練次數
+            * train_batch_size: 訓練資料的batch
+            * convolution_name: Grad-CAM的最後一層的名稱
+            * Number_Of_Classes: Label的類別
+            * Status: 選擇現在資料集的狀態
+            * device: 決定使用GPU或CPU
+
+            ## Method:
+            * processing_main: 實驗物件的進入點
+            * construct_model: 決定實驗用的Model
+            * Training_Step: 訓練步驟，開始進行訓練驗證的部分
+            * Evaluate_Model: 驗證模型的準確度
+            * record_matrix_image: 劃出混淆矩陣(熱力圖)
+            * record_everyTime_test_result: 記錄我單次的訓練結果並將它輸出到檔案中
         '''
 
         self.Topic_Tool = tools
@@ -49,99 +60,72 @@ class experiments():
         self.epoch = 10000
         self.train_batch_size = 128
         self.layers = 1
-        self.convolution_name = self.get_layer_name(self.model_name)
+        self.convolution_name = "block14_sepconv2"
+        self.Number_Of_Classes = Number_Of_Classes
 
         self.Grad = ""
         self.Status = status
+        self.device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 
         pass
 
     def processing_main(self, train, train_label, counter):
         Train, Test, Validation = self.Topic_Tool.Get_Save_Roots(self.Status) # 要換不同資料集就要改
 
-
         start = time.time()
         self.cut_image.process_main(Test, Validation) # 呼叫處理test Data與Validation Data的function
         end = time.time()
         print("讀取testing與validation資料(154)執行時間：%f 秒\n" % (end - start))
 
-        Generator = Image_generator("", "")
+        # Generator = Image_generator("", "")
 
         # 將處理好的test Data 與 Validation Data 丟給這個物件的變數
         self.test, self.test_label = self.cut_image.test, self.cut_image.test_label
         self.validation, self.validation_label = self.cut_image.validation, self.cut_image.validation_label
+        self.Topic_Tool.Set_Zips(train, train_label, "Training")
+        self.Topic_Tool.Set_Zips(self.test, self.test_label, "Testing")
+        self.Topic_Tool.Set_Zips(self.validation, self.validation_label, "Validation")
 
-        self.Grad = Grad_CAM(self.Topic_Tool.Get_Data_Label(), self.test_label, self.experiment_name, self.convolution_name)
+        self.Training_Zip, self.Testing_Zip, self.Validation_Zip = self.Topic_Tool.Get_Zip()
+
+        # self.Grad = Grad_CAM(self.Topic_Tool.Get_Data_Label(), self.test_label, self.experiment_name, self.convolution_name)
 
         cnn_model = self.construct_model() # 呼叫讀取模型的function
+        step = All_Step(self.Training_Zip, self.Testing_Zip, self.Validation_Zip, cnn_model, self.epoch, self.Number_Of_Classes)
             
             # model_dir = '../save_the_best_model/Topic/Remove background with Normal image/best_model( 2023-10-17 )-2.h5' # 這是一個儲存模型權重的路徑，每一個模型都有一個自己權重儲存的檔
             # if os.path.exists(model_dir): # 如果這個檔案存在
             #     cnn_model.load_weights(model_dir) # 將模型權重讀出來
-            #     print("讀出權重\n")
-
-        Optimizer = SGD(learning_rate =  0.045, momentum = 0.9) # 決定優化器與學習率
-
-        cnn_model.compile(
-            loss = "binary_crossentropy", 
-            optimizer = Optimizer, 
-            metrics=
-                [
-                    'accuracy', 
-                    keras.metrics.Precision(name='precision'), 
-                    keras.metrics.Recall(name='recall'),
-                    keras.metrics.AUC(name = 'AUC'),
-                ]
-            )  
-                     
-        train_data = Generator.Generator_Content(5) # 叫入ImageGeneratorr的物件，為了要讓我訓練時可以分批次讀取資料，GPU記憶體才不會爆
-        cnn_model.summary() # 顯示模型架構
+            #     print("讀出權重\n") 
         print("\n\n\n讀取訓練資料(70000)執行時間：%f 秒\n\n" % (end - start))
-        history = cnn_model.fit(
-            train_data.flow(train, train_label, batch_size = self.generator_batch_size),
-            # x = train,
-            # y = train_label,
-            epochs = self.epoch, 
-            batch_size = self.train_batch_size,
-            validation_data = (self.validation, self.validation_label),
-            callbacks = call_back(self.experiment_name, counter) # 呼叫 call back list
-            # callbacks = call_back("best_model", self.counter) # 呼叫 call back list
-        )
 
-        Matrix = self.record_matrix_image(cnn_model, self.experiment_name, counter) # 紀錄混淆矩陣的function
-        loss, accuracy, precision, recall, AUC = cnn_model.evaluate(self.test, self.test_label) # 預測結果
+        train_losses, val_losses, train_accuracies, val_accuracies = step.Training_Step(self.model_name, counter)
+        loss, accuracy, precision, recall, AUC, f1, True_Label, Predict_Label = step.Evaluate_Model(cnn_model)
 
-        # 防分母為0的時候
-        if recall == 0 or precision == 0:
-            f = 0
-        else:
-            f = (1 + 0.5 * 0.5) * ((recall * precision) / (0.5 * 0.5 * recall + precision))
-                
-        print(self.record_everyTime_test_result(loss, accuracy, precision, recall, AUC, f, counter, self.experiment_name, Matrix)) # 紀錄當前訓練完之後的預測結果，並輸出成csv檔
+        self.record_matrix_image(True_Label, Predict_Label, self.model_name, counter)
+        print(self.record_everyTime_test_result(loss, accuracy, precision, recall, AUC, f1, counter, self.experiment_name)) # 紀錄當前訓練完之後的預測結果，並輸出成csv檔
             
-        plot_history(history, "train" + str(counter), self.experiment_name) # 將訓練結果化成圖，並將化出來的圖丟出去儲存
-        self.Grad.process_main(cnn_model, counter, self.test)
+        Losses = [train_losses, val_losses]
+        Accuracyes = [train_accuracies, val_accuracies]
+        plot_history(self.epoch, Losses, Accuracyes, "train" + str(counter), self.experiment_name) # 將訓練結果化成圖，並將化出來的圖丟出去儲存
+        # self.Grad.process_main(cnn_model, counter, self.test)
 
-        return loss, accuracy, precision, recall, AUC, f
+        return loss, accuracy, precision, recall, AUC, f1
 
     def construct_model(self):
         '''決定我這次訓練要用哪個model'''
-        xception = Xception(include_top = False, weights = "imagenet", input_shape = (512, 512, 3))
-        Flatten = GlobalAveragePooling2D()(xception.output)
-        output = Dense(units = 1370, activation = "relu", kernel_regularizer = regularizers.L2())(Flatten)
-        output = Dropout(0.6)(output)
-        output = Dense(units = 2, activation = "softmax")(output)
+        cnn_model = ModifiedXception()
 
-        cnn_model = Model(inputs = xception.input, outputs = output)
+        if torch.cuda.device_count() > 1:
+            cnn_model = nn.DataParallel(cnn_model)
 
+        cnn_model = cnn_model.to(self.device)
         return cnn_model
 
-    def record_matrix_image(self, cnn_model : Model, model_name, index):
+    def record_matrix_image(self, True_Labels, Predict_Labels, model_name, index):
         '''劃出混淆矩陣(熱力圖)'''
-        result = cnn_model.predict(self.test) # 利用predict function來預測結果
-        result = np.argmax(result, axis = 1)  # 將預測出來的結果從one-hot encoding轉成label-encoding
-        y_test = np.argmax(self.test_label, axis = 1)    
-        matrix = confusion_matrix(result, y_test, labels = [0, 1]) # 丟入confusion matrix的function中，以形成混淆矩陣
+        # 計算混淆矩陣
+        matrix = confusion_matrix(True_Labels, Predict_Labels)
         Confusion_Matrix_of_Two_Classification(model_name, matrix, index) # 呼叫畫出confusion matrix的function
 
         return matrix.real
@@ -163,37 +147,4 @@ class experiments():
         File.Save_CSV_File("train_result", Dataframe)
         # File.Save_TXT_File("Matrix_Result : " + str(Matrix), model_name + "_train" + str(indexs))
 
-        return Dataframe
-
-    def get_layer_name(self, model_name):
-        if(self.validation_obj.validation_string(model_name, "VGG19")):
-            return "block5_conv4"
-        if(self.validation_obj.validation_string(model_name, "ResNet50")):
-            return "conv5_block3_3_conv"
-        if(self.validation_obj.validation_string(model_name, "Xception")):
-            return "block14_sepconv2"
-        if(self.validation_obj.validation_string(model_name, "DenseNet121")):
-            return "conv5_block16_concat"
-        if(self.validation_obj.validation_string(model_name, "InceptionResNetV2")):
-            return "conv_7b"
-        if(self.validation_obj.validation_string(model_name, "InceptionV3")):
-            return "conv2d_93"
-        if(self.validation_obj.validation_string(model_name, "MobileNet")):
-            return "conv_pw_13"
-        if(self.validation_obj.validation_string(model_name, "MobileNetV2")):
-            return "Conv_1"
-        if(self.validation_obj.validation_string(model_name, "NASNetLarge")):
-            return "separable_conv_2_normal_left5_18"
-        if(self.validation_obj.validation_string(model_name, "ResNet101")):
-            return "conv5_block3_3_conv"
-        if(self.validation_obj.validation_string(model_name, "ResNet101V2")):
-            return "conv5_block3_3_conv"
-        if(self.validation_obj.validation_string(model_name, "ResNet152")):
-            return "conv5_block3_3_conv"
-        if(self.validation_obj.validation_string(model_name, "ResNet152V2")):
-            return "conv5_block3_out"
-        if(self.validation_obj.validation_string(model_name, "ResNet50v2")):
-            return "conv5_block3_out"
-        if(self.validation_obj.validation_string(model_name, "VGG16")):
-            return "block5_conv3"
-
+        return Dataframe
\ No newline at end of file
diff --git a/main.py b/main.py
index 0379c41..c54eefc 100644
--- a/main.py
+++ b/main.py
@@ -6,18 +6,17 @@ from model_data_processing.processing import shuffle_data
 from Load_process.LoadData import Load_Data_Prepare
 from Calculate_Process.Calculate import Calculate
 from merge_class.merge import merge
-import tensorflow as tf
 import time
+import torch
 import os
 
 if __name__ == "__main__":
     # 測試GPU是否可用
-    print('TensorFlow version:', tf.__version__)
-    physical_devices = tf.config.experimental.list_physical_devices('GPU')
-    print(physical_devices)
-    assert len(physical_devices) > 0, "Not enough GPU hardware devices available"
-    tf.config.experimental.set_memory_growth(physical_devices[0], True)
-    os.environ["CUDA_VISIBLE_DEVICES"]='0'
+    flag = torch.cuda.is_available()
+    if not flag:
+        print("CUDA不可用\n")
+    else:
+        print(f"CUDA可用，數量為{torch.cuda.device_count()}\n")
 
     Status = 2 # 決定要使用什麼資料集
     # 要換不同資料集就要改
@@ -86,6 +85,7 @@ if __name__ == "__main__":
 
         print(len(training_data))
         training_data, train_label = image_processing.image_data_processing(training_data, train_label) # 將讀出來的檔做正規化。降label轉成numpy array 格式
+        training_data = image_processing.normalization(training_data)
 
         end = time.time()
         print("\n\n\n讀取訓練資料(70000)執行時間：%f 秒\n\n" % (end - start))
diff --git a/merge_class/__pycache__/merge.cpython-311.pyc b/merge_class/__pycache__/merge.cpython-311.pyc
index 83892d8e80c13946d2438bf3df97ac2b8815debb..be701db5e3114cdc155eead70fbfc2c67c791643 100644
GIT binary patch
delta 999
zcmZ8fO-vI}5PoleO55$SZ7EP7Kx0KM6tSSis9?0!n2;a_4_GmE(=D~81$|3Qz^!t?
z*aI=aci`foi3CYFNYb;3Cr=*G7$uDdFB(ipY$SMg-d6dWWWV{Jc{?-D+Ln*VuVmSd
zfbIr|Qy1su=?0EzKwGx~7Elt!u!drs=A19IhODzFX4PDpeHO=Tz$M^z&5@MGI8B}v
zVh*rXgG~mm0I#0koH@GP3zed0M)k>wteUtJPpIRGBpDqd6Dgu*eVx9I#ituD-xLk}
zj-$15UG-_Rvw;mR#pQS!lD&@&Uc(x<#3Qr<UxkKIfFC66y+0WQa7oD~Nh%rF)JZj-
zRny}Efk@yf*&-QL*VAKDp<S`NB$?5g8LkTu=K~@Rmf)!rQZt!&I;*CVgJ#flqiH^V
zz-jtwAx%FFFoV|XeTBeUT`(_wQ)<2_{*Q{kFtF-cz3@Q^e^SD0)^M>;xPkF+x*&!*
zzrXlijNxYDgXHPjy&ZP`qp^6k>KDfC^t>%-2|^s{r1x!&I6&uYhrs;W_Pl;)EZ`R5
zWBS$~L`UqGuJLSq4PdBH3W3h;)ze%~F!(6CbulNx+{Im=!R^&x9~O<zRndN_0bA&Z
z)N!+F35LJ&|7wsS7~(l-GZ!%gSddUz5IF%R(h2}8X=ieXftx`Gvs%qXky9X-EZHeN
znHeJ?x+%2?RxnY;an{lUf<8xw9Ko)#$UE(Mp=?JUZyAZUJ`?hhb+4~b^Fl5>SWYZH
zT&wTNUs-q8<RkQ}<9pY?=1owg@&YH5x*3?2_Z<M~D3;|Xc%Td10kI?!bv%_M&HHmb
zrnB<!lgw4w3`ixM*4Yl&K~*k_bkny=(bC3J6#B^7-^y;q+&Kv_=@<hxICCBl{@K}r
sf6#4bT;avuGTH$<z3vJ&Rg0eGXc+-J2tHbH-NRApa(nS*+V0l=0@{e^F8}}l

delta 904
zcmZ8fO-vI}5Z<?c-QDh%U4AVnVq4Hcfp|d!NQ{6UjD(06E2b{IrD<r3dDTS8HgJH1
z7-CwV2_8IfQWA+V>BR#k9E?{vHjM`_9z4;6lboElU~D>v@B3!v&6}CGyVLWs+jwsn
z8VB~eh2jTAV>u`=5A0|bV2g9OIKo_<#|_2p^Zh1P*HB!+Cf0yeY+=8nE$~=hlj44e
z8W7ci1HeYHx@sPWCx>UJrpITe$x@QonbH>e#+MeQdk7uUeR-(#N4}4um4=K%l1>8K
zPwYg-PGy%{yud(RxTUnD93m8^2`wgf!+fNN7PWR1qPtoMb<^+Kt2U7he`+!?+vn&e
zqR;%3E21l66#SRvk{tMaoiEh-DIqVp;sp0_E-$+R<`#KYOqyKe%Ky)~e65${^FA!(
zm29y7RH*itE7W>P0gA-DLIZjb`RFCRFQvSLW*@wUtf?!x@+k1;yerLcQE7^>6zK){
z?Ws;sQIUjMw2wg}g9wYN#Lx5~Kt;|x&pGLZif_(N+u3=C^wE!cr=&uziyr7B@&KsZ
z7it-?M`hhCSev&BcdI_p)HcMb&NZK@a;kdGgN^ZWL+hK!i<OO)^4ZAt%-cIVL!ZLC
zBL^K<3QD=PtsoSHpS6k4zI|a&Kj@e|P1CgTYw$AkgSk6^PKGy`iZSR1$gvOF1x9oh
zij-CZ>*#hAbX2NH#Lgxia*k!Z#l6~QFL|;v+lHL{DxfMTm6L$keWw@3=(Jh#v&SO}
zKr}#AYpneu^Fs_SFu2NqJtc|J1*;SNpzBtG9$B%jCaJl~0qjh8R1dv$xbboObygW*
Vz;-xRMl(zJY1$bhH|W<!{1<yo!sh@0

diff --git a/model_data_processing/__pycache__/__init__.cpython-311.pyc b/model_data_processing/__pycache__/__init__.cpython-311.pyc
index ae4bbc70291ec6f58fdd77f97727dc36de34e60d..58b90c4ba1ad51f9bffafe145ecc6eaf4abe5d73 100644
GIT binary patch
delta 65
zcmZ3)ypx${IWI340}v$W1gCH0(PLEq>H_4sSj7~V<mV<PXT&Eb<|U^V#RL@PrxzvW
SD(EV_SknGt@!rW^j1d4<7!{EK

delta 42
wcmdnVyoi}+IWI340}zB6Tu<J}qsJ)l%fQtt#xKA*CZH%ky(lqvaw1~{0PoHWTmS$7

diff --git a/model_data_processing/__pycache__/processing.cpython-311.pyc b/model_data_processing/__pycache__/processing.cpython-311.pyc
index 9da029dd22b25ed9885575c7f5a60118b1d5a4d2..6406322ebc9c704f886932ad12c26af1e606686c 100644
GIT binary patch
delta 605
zcmew+^n;ggIWI340}v$W1gAH#OyrYLk^%CjGo&!2Fy=5sF{CiHFhnt?u%@uIFh((@
zFa|SdvTY2~WmNy_22|%_6;oW2pPQJR5ucoxmz-J@6Ht_&UX+-tpsVm=N&Ab%dnflZ
zo@FeW9Lw}msfICyNfKli7^E`*`HaZ?Os2_4n4{`7nM;`&85p<}fIwMU;l-lW&!?<-
zvajuF<E$5(+n-IF`K)vDlU+^EXSD#uUp7pEDpXc@HmCDx>&oXV);-zL`*cDNP{GSN
z4bL_<ffX3)DM0i+-MjH=b3eo+jTh@0p04Y8wra}LeIWH9x%%dQO$A#!g{QOUJ#E_T
zlbDp6^P;2OGdD3kRUs%pzXWK-(}r%J%;FNTxkfN^Uu<9hbl*g%xe!Z$CL)Wz-_!H5
zspZ+GNkCgcR(AFxSp;?5z8(pnE|<(?i0MUwz_2P}0uq|6lZ{wZCP%a6izooOJV0Fh
z2uL(AeBfi?;F)}frCIg^8v`d_M`91h1fv@qyb#6*R%S`A4&DzOlb5lYx;)^QxxgV)
zBmmT?$p|)FletI~$SdLp5eh(JB?DNxh!4pA#bJ}1pHiBWYFDHL<N{-bq1bw|6sMA+
NG^6YX1{?%fKLAjL;BEi_

delta 1166
zcmah|%}*0S6rb7MZlT?lZM8)f3Q81Qg&0XVP)G_<BalEqIFV}8u%+^4&9oRzwkc5#
ztKraQiS*DzPkuxXdNA?ie_&Eb6PhL@UOec<HAbVJe6y`eA?VA@Z{NIm^WK}^y#4O@
z=5{`IIBW={^;2sk%4eL%x;kLBi4>&ZB{YK+W)00?g;%&WJi|oM99Ar#u?n6-vi0R~
zHx_^PT=ECUCoTjhNGwXi%X$vK#c%Wl=0_@Fr?D%d;zt;vXD}<G;6;XN#I9n6X{uTg
zilTX(WN%3Ot+2*>y<pAmUC*;gR>c?5)y||<MXD9{^oQ)o(Wl%vanU?9+Xo6VuYcuE
zS_lui-eifn+d&av#%)mDDk@p+dd_n1@_V~eTe<IF_h(%X0$Sjm=t+k+>asls%lMqy
zm^NnEki(rmu|Aw8CBd~%F{~W*GHm()-<v8mv=r)EhzMC`Ie_0nYhYo1xelo}5xr!U
zmRpHGR3W6Yhmyx0Adds`_F1oip_V=N0DGLu0(<>QOM(_u6&2Mc1`j#OtDI{2%YB~F
zKQo>NX0J`<^#I%4cV>NnzEkx|=b|Xd{U^GQMz2I<nM#yYtIACwas*0}S@YG11$06y
zkga7)d^WrkDccrSlxQTh98)4?P6;Q%q#5?hHgh*b4dwboED>G`Ek^DT2kgbI0LLpr
zoYdJJzZoXsL?kpfe><X>hlq?)kcPT{1fbpFN>AQRt)_Ns+oL44<#FzG{z|q@8`VY)
z{@=YJIl)IjQk#L8`XO9FZDiLYWqUKhji7<BgPH+TIE_o9l<g~ur*_1yZL#ag^b7fU
z==Ef=JFq1V7RAAWISkf3!=pNPrNok_QM2v3yy=xp%SH<XtM4>H8!8Fj9l^IP__9j=
zd@i~rbQFb-f;pga!CT-wX0ytexV&(0nInzxAq^Dh|3YLv!%Mcapwd2#PwTJvcE5yu
S2C4-=>gzX<2YU|%Gtn<EYy$`Y