From 11b8eb53a547afedea7611a5a5f9fdb190162451 Mon Sep 17 00:00:00 2001
From: haotian <2421912570@qq.com>
Date: Thu, 10 Apr 2025 14:16:46 +0800
Subject: [PATCH] =?UTF-8?q?=E4=BF=AE=E6=94=B9--=E4=BF=AE=E6=94=B9=E6=96=87?=
 =?UTF-8?q?=E6=A1=A3?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 doc/典型示例新版.md                           |  14 +++++++++++---
 doc/接口文档code.md                           |   9 ++++++++-
 .../__pycache__/data_manager.cpython-39.pyc   | Bin 18254 -> 18260 bytes
 function/data_manager.py                      |   7 +++++--
 4 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/doc/典型示例新版.md b/doc/典型示例新版.md
index b512d97..6725480 100644
--- a/doc/典型示例新版.md
+++ b/doc/典型示例新版.md
@@ -316,7 +316,8 @@ TEP数据集共包含53个特征列：
 1. **完整性检查**：确认数据集无缺失值
 2. **异常值处理**：使用IsolationForest方法识别并移除异常值
 3. **数据标准化**：采用StandardScaler方法将特征标准化到[-1, 1]区间
-4. **数据集划分**：按8:2比例划分训练集和测试集
+4. **特征提取**: 使用PCA将特征维度从53降低到20
+5. **数据集划分**：按8:2比例划分训练集和测试集
 
 **预处理配置**：
 ```json
@@ -327,7 +328,7 @@ TEP数据集共包含53个特征列：
         {
             "method_name": "IsolationForest",
             "params": {
-                "n_estimators": 100,
+                "n_estimators": 200,
                 "max_samples": "auto"
             }
         },
@@ -339,7 +340,14 @@ TEP数据集共包含53个特征列：
             }
         }
     ],
-    "feature_methods": [],
+    "feature_methods": [
+        {
+           "method_name":"PCA",
+           "params":{
+                "n_components": 20
+            }
+        }
+    ],
     "split_params": {
         "train_size": 0.8,
         "val_size": 0.2
diff --git a/doc/接口文档code.md b/doc/接口文档code.md
index 2b9a739..f2e319e 100644
--- a/doc/接口文档code.md
+++ b/doc/接口文档code.md
@@ -151,7 +151,14 @@ Request:
             }
         }
     ],
-    "feature_methods": [],
+    "feature_methods": [
+      {
+           "method_name":"PCA",
+           "params":{
+            "n_components": 20
+         }
+      }
+    ],
     "split_params": {
         "train_size": 0.8,
         "val_size": 0.2
diff --git a/function/__pycache__/data_manager.cpython-39.pyc b/function/__pycache__/data_manager.cpython-39.pyc
index c18e45b2cc126207487287f93e6c90bfb715ca65..c1e1bafd78477cb353edd0b53bc2d1faa896f8a3 100644
GIT binary patch
delta 1014
zcmah|-Afcv6u;-r?9S}R%=nROiz0ru*%qvzk|;$fC`Cc|P)Q=yb+yq(8&@g2<CeR!
zBlxj&38IL0-^zzb#)qCF3VM_vdyW2!z^Cran&m?{Fz3$wUC!_KJ7><_UW1)AP;M$p
z5b@`J$FH&Tn@WG(h{ks=B=C1;3nP@r^Wzi~h$2g%xPS=0gNoP^Dd<21A}(Qm;`2^?
z)K|pIhzb<xcn-`9DS`N@@B+QW8CV1tWt(-CX&@rc@pA@|GYCt##FKQHBns5eSD6Y#
znnQUo>#1a;C@T1h$O>>&p1|fO8}*^#h!wPP!A}D;7)3>J*B*55wb4^Ti?9qS$%d4y
z2Qt{)O$GIk4LOv~K@yVy31(%NCIeK`!z8pSEQ1YPLedo*k?<;55>HX7L<3`(C`4Vw
zmXrshk(<ywYtK9(&t4I%kd1YXlucl6`dSVrft3H%u)>B)D)=7RX;0J&yMAbhhX2SP
zCT(nR;Yv?<_n<Hh>19OWT~kV&<cKSjDyK>Y{w=Vikp6#l#=t@1J-RB-_kms=W_GW0
z@_y&s-(OLvBoWF!)<=5Z0<u&<^CvvGid4IELXM^)RLuaN7_kU|7SUC79TCA&7gQry
zS`#S2kwMfNWzmM1eAyi4lWMN9D-EseQNw$fW?hXJC3D?H^Xzfs3Rp~SnuIia*t7ua
z{QE5HYL3HmHrhPOzBc#!heTpdO|$Bjwqu&v>E`9!Fz$w=8^(AT3X<f=?WxJxyT%N=
z)iR|~4k9dq3}LL|u=As(9xxT3#A;6Ys7!=eFpZQz<qR17_VC*HoO)=Mqk(-?Vgs>m
zkXbeshX8vQi?g4xZ?NHP9ls&Kj#GZ}vjCWB?fp(|Vg(@KylET7aLLiy;{ep@?pOqP
n>U`)N0LVDf=`VhG<!qnHVi29dbF~<1oatTyFyeggdw_ofWEcdD

delta 1038
zcma)5T}TvB6u#%q?(EF&&ie1V8saaTt)fDSkQAbz81y0(hCUS6HOE?O+$5JfZn?_t
zLas{}3i^RvkOB#!<|XQ-2<j~;q!)Xu9(wP=w>q<HTS8C+bI!RR-0%C&;oR*Eyvu-g
zR@2-Bzxs{uGu|iKNMqcGCmWLRUwBFh$<kPuQw9NIN}T2hqq8JWO-G@YIOxC}%*o_+
zs?$786E1P0ON(HPMjgz>rMu)lO~4A+D3<~xOlAppY(fUo6mL_eAhQ!ICsUakU!$0m
zH5#S}%N)29Gtw~kBFTc$$mJ9Xker++rh=$vjv6mhWSC6Vnre#XTwLL5faJl>1LX39
z$UqUzLmHw=3Zkm+PEcbv>e0O^NRlLkDf2RIRk2wTap1D9G2gnBh7_QLtV4>hvUM<J
zyyD|BS7vCjUsKLPqy<fs7p&bX2o>z*!E~pnUdAfsD4IT%q+wu{B@r+^5g)6<GbxDT
z*|x*o+!OyL|1ast77AAvgm?D~bGKg3?5TN+q^!mkidJxM0<gnUgMU?+|D(Ie=ziio
zy4v5q|JUr^<Xol6+1pT=t73lfs>wg}2#Bcxi5)7SHLT9w8WxDwaxDS4KaM;ASs|y%
z1YxSFEo%{XG53%jM8-)+y=Z6-Ru~yvTH}r#HtdiUW6d41F@^O>vC#YsE{e{U6XJSH
z8e+oJdI7k&+PVxkus$yO+Cp$!oNJqw&p4Pde@VP;8xhs*z4bn$&n8RS;k+Hfc9_8+
z=Gy1A7#4&&Kmjk*S>M{50M|o!f(>1`DonZpMkFe6B>@q9MA(*#s|9owQMH6BVmx>h
zRIw2Zfm^%?hN!V6J_kR-L+km$NeSLt{?6AD{Q=ge?hgQ6*5lr33S(lYdqhkg{sK$Z
z;E@&Z+{6cKY@(r9Nc8smt*?DSfNPf4|3-#;*7mVg3SdnP9-wG-afks-T06r_^d~5^
B3WNXv

diff --git a/function/data_manager.py b/function/data_manager.py
index 6be6719..daf5611 100644
--- a/function/data_manager.py
+++ b/function/data_manager.py
@@ -336,7 +336,10 @@ class DataManager:
         try:
             method_name = method['method_name']
             params = method.get('params', {})
-            columns = method.get('columns', df.drop('target', axis=1).columns)  # 排除target列
+            
+            df_columns = df.columns[:-1]
+            
+            columns = method.get('columns', df_columns)  # 排除target列
             
             if method_name not in self.feature_engineering_methods:
                 raise ValueError(f"Unknown feature engineering method: {method_name}")
@@ -445,7 +448,7 @@ class DataManager:
                 train_data, val_data = train_test_split(
                     train_val_data,
                     test_size=val_size_adjusted,
-                    random_state=42
+                    random_state=random_state
                 )
             else:
                 train_data = train_val_data