From 672b4524a9fba7bb0ff8721daefbc8ffefc91d9a Mon Sep 17 00:00:00 2001
From: saeedkhosravi94 <saeedkhosravi72@gmail.com>
Date: Sat, 6 Dec 2025 21:52:40 +0100
Subject: [PATCH] training data without x features

---
 .../catboost_model.cpython-312.pyc            | Bin 7979 -> 8069 bytes
 .../lightgbm_model.cpython-312.pyc            | Bin 8582 -> 8672 bytes
 models/catboost_model.py                      |  14 ++++++++++++--
 models/compare_models.py                      |   8 ++++----
 models/lightgbm_model.py                      |  14 ++++++++++++--
 5 files changed, 28 insertions(+), 8 deletions(-)

diff --git a/models/__pycache__/catboost_model.cpython-312.pyc b/models/__pycache__/catboost_model.cpython-312.pyc
index f0c02c86d01b094b005192a2f605cf577489220d..5823aaae6838634084201eb9e1b8dab2a16141d3 100644
GIT binary patch
delta 761
zcmYjO-%Aux6ux(NW_H$fcXV@ho!wlQz+if?SywADR>~6VMbdqUwV2F1Ya^yMJ1egF
z5M*E^J#>Q#`Uk>#h>6JO=p~T~l$I&Thajwn%!QzbpmR6U4&0CPogd#hb7m=XqdT+`
z4Avm93pd-%?xv@qi`d{92df+bxQ7gs!vy6#CL)-4h?j7Ap78TH=Ow~C%5lU8$ddr^
zLrfqwz<ec9t#$o9$FlnI{{F%2CX*aZSM6o@<^f>zjgp9t_`v&T95`Nhm7mU6yy+3<
zggK-)&=qbK%k%;NFye+Mi}qB}*7N4Lq1(j?b7EBW(ZfP>z&Z$x1*VpyLqa~FVuY@t
ztuTtWIIo2$4y=jsjeYG4+`8CIKl{GdYOEfLP|`}#Phtz!DDOXo(@wwtIgh)jAl<@U
z&K;=_L*_%>vXX*YYXD$2xbs=9dDA}RJgJ}Y;56Oe@FO+|c{fnfr!0f$wo$MxKSWf{
z(r8L<i8BqCGp2{fXG6Opj9;Z!<RK7Ql@<Ka`64fSxH&YY(KFsX_44D$k_S4$Fe)K6
zN}o0M;a>W?QO6m2UKwthf|SL6mWo|&c*ZRBg2S5=#F$YP`b#+(=wKCWdn-#%#nN&g
zxUOJN>LZ2e>h08sr8$eSAWr^W60U3fza-<=p{cO$EXQAA`F_KDMSG`cFZj1&<C>V+
z5JOg)-fn6gYhzP(=UrlS0pJm`(#)?OQeqy1AiM$#&7&<5)rHHEXgM4y2c>envmzz=
zj_nhe<Fk0Xwg>ZVmD3)9-&?6Whj|IcB@&%@nx0LRVv9^~N8S~WUY;byaU)~hrQZ@M
NX~2u6ZN#uz^aohTz%Kv*

delta 677
zcmYjN&rcIU6rQ);U3RH$Az2DtN=u@ctf~Dafq;lXQ>|Qx(x|D2wRE<m2~yqNl9rQ+
z5D%Ql9K0B#@nrC#n<$(Kfy7G>nqW^HJa{w0AE0m0h%b5HydU5D=FO~L-k1!23IsX{
zny@-teGz>ZoP+fNbz%f7&xl2eKuJ-k5(-qHzKTf26)5_sv_gs^l@TS{NfoTgRK={L
z*<twEM~FqsG8g@8BA!0<qwfNjWiPb>FCpv-45S^<*sk<kFl5%0<C4S2*lW3>#(|K>
zWM4nyX&t@b`=a!-5#`6J6!sl#kwH7nGU^bVWp~sYko8`vA0!xOcSCD1?tKg80aEOI
z_f{m0T>B=%a^UJz!MrjvGh0Yjs&2yj)!h&v%Np8G{TBAy{-*iRwy5b^j%%w}F+@AX
zp7ad$bB(rWEtM9QgC`<9f5f(X?x9FL9DzM=EW9OXw1<Y@_HN)yxdCYsqmSv4Jmgq8
zV!|c19VuO`VapyxFrY1#8dYaGS}wWfT&?D~rdwaEF3y{_<<u8kCs}rut|3jTE>f$3
zT2p$?Kg{R(tL#~}u3zZ7j$vDNYbKJ7u$}V0>H!%1vp33F|L@K7Ci0z@Cr5Xns}&jD
zthbcD&r0q<3EEk9Fp!+iaP5;@=6U)Q;SI5~JU;%`5RW(n&>aI|h3ub8Y)<WIt-|di
y5|&OMrrG^iFaPG)7}VMOSW{o)Q~yNT%EW^jtuI)UHe<PXCY1L<=#X$cCjAE3{HnVE

diff --git a/models/__pycache__/lightgbm_model.cpython-312.pyc b/models/__pycache__/lightgbm_model.cpython-312.pyc
index 3e9e0629b06958db21a2885906091fd7064298f5..bddacf0e1ae3dce53ee70346106478a867fe0cf8 100644
GIT binary patch
delta 802
zcmYjO&rcIU6y9lfw@a7pmX&UyTcuzc-3x9h1Pp|OB&8&ncpyZp$x_)ZF7yYP-4Y<x
z7()!=$xJjxy%VDsvj=}9T)dcQOfbRN?1{fXu{|0SXNrjL@a28)ee=CHGj}KUCqkbD
zAxPo)^=Qz#)cY!w1wn>z9L*6l2O?(;y1*FB2FUphc7w{%MgSvY@J1W1u|^QHz#6Bv
zqo?51;JQ+@7H&8T*GsxnEnDRU9h!Ev=-B=9_R4(V(71&(Mn9P%xx0Yzvo*Fx=`#9E
zzXMUU$?V9URH-`4RYxyaMf0zl8bFh54-e1bgEnzXpe44z4+Dz2O&v&-l5{;b3HUuu
z+3y?Nq`z`K$jAL?*N7gAs02)*-&_(5pwn$rAnlgg;w%_NTjG6;a`<Lv28+D*vToXr
zRmwXRXv0x=J1hdAq1PR6yHmIiFJo92rl%&eGy3F(i~R-58FueU+dhy+smRaR6=K4#
z>C4bGbO&2O4p&qfYG_POD#XL1&6Rv{J#=hB_#IS}uPSNWpC{WMMF&n=Z>6~nS@44z
z$_o+6kY7e$_zpFITBFo1bRw1o8FVc+2gcD(tkSoFyKsa+Ck7RGbbi&cGdN+avSF^O
zakQk&@IypFE(LSwqmqtH<J4oFijF_6_10<}YOWA}0uqP2<G(fk-<|L+e8@iIemEKR
zMILv4>(;(>Yj2ocu4|7Q-{(RwjlTEvFAk6?$MedBKEUvtg5=te_`}c<2sp7xX@ff8
zsIXL*qjgEH3u0XvX^MT!P$LWc%nWD*rvNk1ob|CxycsS5M#ST8CC>9z;xtR3hTbO@
gV-HFCZ;;20l`2NHXpX~YDASt~7yUqNPy}0|Ka|(US^xk5

delta 687
zcmYjNOKTHR7@eDWWTtr<(mbbi+8S}hHf;<Qi;5a+Q(qh7t1gm|nMkBb<j$ld6`@EF
zK?LQxQUrG*3StMLB8u(=5wcJS<EH<h*lxP?yQx|~xM#j|&*98>ztzm0jQ6X@)5kz<
zw$qi@${TMMJsgnGBq^Z&-DOVW+Gvi|_%<`gX#yax^=Ts937Q1V*>b8b@)Qk*O9Bn~
zQyhF*I2)X`hIp1Z_+=C#Z}|@nRUmhToxDNaWLYRnqljUam~L#BHR4Pl_J<fJ=fyR5
z5_AJ3M#1CcoRmO&h$+pYl=V@P1vEu|`tAW<^WO?igUNQ+^15MGsztMo4LoIS`h5r`
zi5Pen9S0p>0$B8%nVrez@-xTJj+HCssAUJ9J5Y+;?*A3LOha6){0*#Yc@v_r2sbK+
zljMFV5vLh8r!N%e7romdlsCz@(A9Vf^b(!+Z|#7lw*cA7ggtCf3u)n3k(Z7w9WgCN
zjgn?8i>ArvSOFa-!|{5k0V+N~Aqb!%n=3Xe26>ML+$VwYm71nE)i|-@Iaiw6=&SG*
z5|H=%PeW@9W<6gr7E~E0tvUHIQnuzc8#Vp^xhY=)zcFkD2g443Cpx*Jcf`nd@x-d=
z#VK-B89OyWv$y|JrSt*73kK8I;@(=lw4>mGW2gXarn`6F%CRrOPVV9c6X8eJ)8vy9
whBbGUgD#W$FHpEi<e}@aN3{JHXN!lbb*(Y4AHh$^v!MxJfknP`hGNh38|w(ADF6Tf

diff --git a/models/catboost_model.py b/models/catboost_model.py
index c312428..a14ce71 100644
--- a/models/catboost_model.py
+++ b/models/catboost_model.py
@@ -11,7 +11,15 @@ from tqdm import tqdm
 
 
 class CAT_BOOST:
-    def __init__(self, data_frame, params={}, n_split_kfold=5, test_size=0.15, seed=42):
+    def __init__(
+        self,
+        data_frame,
+        params={},
+        n_split_kfold=5,
+        test_size=0.15,
+        seed=42,
+        output_file_tuning="cat_boost_tuning_results.csv",
+    ):
         self.data_frame = data_frame
         self.params = params
         self.n_split_kfold = n_split_kfold
@@ -35,6 +43,8 @@ class CAT_BOOST:
         self.kmeans_estimator = self.params.get("kmeans_estimator", 5)
         self.tuning_results = None
 
+        self.output_file_tuning = output_file_tuning
+
     def preprocess(self):
         self.scaling_method = self.params.get("scaling_method", None)
 
@@ -208,6 +218,6 @@ class CAT_BOOST:
         df_tuning = pandas.concat(
             [df_tuning.drop(columns=["metrics"]), metrics_df], axis=1
         )
-        df_tuning.to_csv("cat_boost_tuning_results.csv", index=False)
+        df_tuning.to_csv(self.output_file_tuning, index=False)
 
         return
diff --git a/models/compare_models.py b/models/compare_models.py
index dc5e157..5d1142a 100644
--- a/models/compare_models.py
+++ b/models/compare_models.py
@@ -2,9 +2,9 @@ import pandas
 from catboost_model import CAT_BOOST
 from lightgbm_model import LIGHT_GBM
 
-data_frame = pandas.read_csv("./data/Ketamine_icp_no_missing.csv")
+data_frame = pandas.read_csv("../data/Ketamine_icp_no_missing.csv")
 
-cat_boost_results = pandas.read_csv("./cat_boost_tuning_results.csv")
+cat_boost_results = pandas.read_csv("./cat_boost_tuning_results_no_x.csv")
 lgbm_results = pandas.read_csv("./lightgbm_tuning_results.csv")
 
 
@@ -44,12 +44,12 @@ lgbm_test_metrics_clean = clean_metrics(lgbm_test_metrics)
 
 comparison_df = pd.DataFrame(
     [
-        {"model": "catboost", **cat_test_metrics_clean},
+        {"model": "lightgbm_no_x", **cat_test_metrics_clean},
         {"model": "lightgbm", **lgbm_test_metrics_clean},
     ]
 )
 
-comparison_filename = "comparison_catboost_lightgbm.csv"
+comparison_filename = "comparison_lightgbm_no_x_vs_lightgbm.csv"
 comparison_df.to_csv(comparison_filename, index=False)
 
 print(f"Comparison saved to: {comparison_filename}")
diff --git a/models/lightgbm_model.py b/models/lightgbm_model.py
index 5b3dd59..0d27a8a 100644
--- a/models/lightgbm_model.py
+++ b/models/lightgbm_model.py
@@ -11,7 +11,15 @@ from tqdm import tqdm
 
 
 class LIGHT_GBM:
-    def __init__(self, data_frame, params={}, n_split_kfold=5, test_size=0.15, seed=42):
+    def __init__(
+        self,
+        data_frame,
+        params={},
+        n_split_kfold=5,
+        test_size=0.15,
+        seed=42,
+        output_file_tuning="lightgbm_tuning_results.csv",
+    ):
         self.data_frame = data_frame
         self.params = params
         self.n_split_kfold = n_split_kfold
@@ -37,6 +45,8 @@ class LIGHT_GBM:
         self.kmeans_estimator = self.params.get("kmeans_estimator", 5)
         self.tuning_results = None
 
+        self.output_file_tuning = output_file_tuning
+
     def preprocess(self):
         self.scaling_method = self.params.get("scaling_method", None)
         if self.scaling_method:
@@ -232,6 +242,6 @@ class LIGHT_GBM:
         df_tuning = pandas.concat(
             [df_tuning.drop(columns=["metrics"]), metrics_df], axis=1
         )
-        df_tuning.to_csv("lightgbm_tuning_results.csv", index=False)
+        df_tuning.to_csv(self.output_file_tuning, index=False)
 
         return