如何解决使用 chi2 的 SelectKBest() 的 K 值
我在这里使用了稍微修改过的代码：Ensemble Methods: Tuning a XGBoost model with Scikit-Learn
当我执行它时，我不断收到此错误：
ValueError: k should be >=0,<= n_features = 4; got 10. Use k='all' to return all features.
我有四个功能和一个目标。我已经在 SeleckKBest() 函数的 Pipeline 参数中的下面代码中为 k 尝试了值 1-4，但同样的错误仍然存在。
这是我的可重现代码：
import pandas as pd
df = pd.DataFrame({'Number1': [11,12,13,14,15,16,17,18,19,20],'Color1': ['Red','Blue','Green','Yellow','Orange','Red','Orange'],'Number2': [221,222,223,224,225,226,227,228,229,230],'Trait1': ['Jogger','Sedentary','Tennis_Player','Graveyard','Shift_Worker','Jogger','Fulltime_Mom','Couch_Potato','Graveyard_Shift_Worder'],'Target': ['yes','no','yes','no']})
col = pd.Categorical(df['Target'])
df['Target'] = col.codes
from sklearn.base import BaseEstimator,TransformerMixin
from sklearn.preprocessing import MinMaxScaler
class PreprocessTransformer(BaseEstimator,TransformerMixin):
def __init__(self,cat_features,num_features):
self.cat_features = cat_features
self.num_features = num_features
def fit(self,X,y=None):
return self
def transform(self,y=None):
df = X.copy()
# Convert columns to categorical
for name in self.cat_features:
col = pd.Categorical(df[name])
df[name] = col.codes
# Normalize numerical features
scaler = MinMaxScaler()
df[self.num_features] = scaler.fit_transform(df[self.num_features])
return df
from sklearn.model_selection import train_test_split
# Split the dataset into training and testing
X_train,X_test,y_train,y_test = train_test_split(
df.drop('Target',axis=1),df['Target'],test_size=0.2,random_state=42,shuffle=True,stratify=df['Target']
)
from sklearn.pipeline import Pipeline
from sklearn.feature_selection import SelectKBest,chi2
import xgboost as xgb
# Get columns list for categorical and numerical
categorical_features = df.select_dtypes('object').columns.tolist()
numerical_features = df.select_dtypes('int64').columns.tolist()
# Create a pipeline
pipe = Pipeline([
('preproc',PreprocessTransformer(categorical_features,numerical_features)),('fs',SelectKBest(k=0)),('clf',xgb.XGBClassifier(objective='binary:logistic'))
])
from sklearn.model_selection import KFold,GridSearchCV
from sklearn.metrics import accuracy_score,make_scorer
# Define our search space for grid search
search_space = [
{
'clf__n_estimators': [50,100,150,200],'clf__learning_rate': [0.01,0.1,0.2,0.3],'clf__max_depth': range(3,10),'clf__colsample_bytree': [i/10.0 for i in range(1,3)],'clf__gamma': [i/10.0 for i in range(3)],'fs__score_func': [chi2],'fs__k': [10],}
]
# Define cross validation
kfold = KFold(n_splits=8,random_state=42)
# AUC and accuracy as score
scoring = {'AUC':'roc_auc','Accuracy':make_scorer(accuracy_score)}
# Define grid search
grid = GridSearchCV(
pipe,param_grid=search_space,cv=kfold,scoring=scoring,refit='AUC',verbose=1,n_jobs=-1
)
# Fit grid search
model = grid.fit(X_train,y_train)
错误：
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-84-5c30ee0bb39f> in <module>
28 )
29 # Fit grid search
---> 30 model = grid.fit(X_train,y_train)
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/model_selection/_search.py in fit(self,y,groups,**fit_params)
737 refit_start_time = time.time()
738 if y is not None:
--> 739 self.best_estimator_.fit(X,**fit_params)
740 else:
741 self.best_estimator_.fit(X,**fit_params)
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/pipeline.py in fit(self,**fit_params)
348 This estimator
349 """
--> 350 Xt,fit_params = self._fit(X,**fit_params)
351 with _print_elapsed_time('Pipeline',352 self._log_message(len(self.steps) - 1)):
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/pipeline.py in _fit(self,**fit_params)
313 message_clsname='Pipeline',314 message=self._log_message(step_idx),--> 315 **fit_params_steps[name])
316 # Replace the transformer of the step with the fitted
317 # transformer. This is necessary when loading the transformer
~/anaconda3/envs/python3/lib/python3.6/site-packages/joblib/memory.py in __call__(self,*args,**kwargs)
353
354 def __call__(self,**kwargs):
--> 355 return self.func(*args,**kwargs)
356
357 def call_and_shelve(self,**kwargs):
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/pipeline.py in _fit_transform_one(transformer,weight,message_clsname,message,**fit_params)
726 with _print_elapsed_time(message_clsname,message):
727 if hasattr(transformer,'fit_transform'):
--> 728 res = transformer.fit_transform(X,**fit_params)
729 else:
730 res = transformer.fit(X,**fit_params).transform(X)
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/base.py in fit_transform(self,**fit_params)
572 else:
573 # fit method of arity 2 (supervised transformation)
--> 574 return self.fit(X,**fit_params).transform(X)
575
576
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/feature_selection/_univariate_selection.py in fit(self,y)
346 % (self.score_func,type(self.score_func)))
347
--> 348 self._check_params(X,y)
349 score_func_ret = self.score_func(X,y)
350 if isinstance(score_func_ret,(list,tuple)):
~/anaconda3/envs/python3/lib/python3.6/site-packages/sklearn/feature_selection/_univariate_selection.py in _check_params(self,y)
512 raise ValueError("k should be >=0,<= n_features = %d; got %r. "
513 "Use k='all' to return all features."
--> 514 % (X.shape[1],self.k))
515
516 def _get_support_mask(self):
ValueError: k should be >=0,<= n_features = 4; got 10. Use k='all' to return all features.
解决方法
有 4 个特征（Number1、Color1、Number2、Trait1）。
SelectKBest 会从原始集合中选出 K 个最能解释的特征，所以 K 应该是一个大于 0 且小于或等于总数的值功能。
您在此行中将 GridSearch 对象设置为始终使用 10：
'fs__k': [10]
在声明期间覆盖您的定义
('fs',SelectKBest(k=0)),您可以删除 fs__k 行并将声明行更正为您想要的 k，或者在 k 定义中设置您想要的 search_grid。
版权声明：本文内容由互联网用户自发贡献，该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容， 请发送邮件至 dio@foxmail.com 举报，一经查实，本站将立刻删除。