交叉验证(Cross Validation)和网格搜索(Grid Search)在python sklearn里的应用

Theory

• clearly explain the theory of grid search and cross-validation, with an example about DecisionTree 决策树预测乳腺癌
  ML 入门：交叉验证与网格搜索算法

sklearn - GridSearchCV

• offical document about extimators
  Developing scikit-learn estimators
• [recommand] theroy with code!!!
  Python机器学习笔记 Grid SearchCV（网格搜索）
• clear introduce the GridSearchCV parameters
  • python机器学习库sklearn——参数优化（网格搜索GridSearchCV、随机搜索RandomizedSearchCV、hyperopt）
  • sklearn-GridSearchCV,CV调节超参使用方法
• also a clear inroduction
  python机器学习库sklearn——参数优化（网格搜索GridSearchCV、随机搜索RandomizedSearchCV、hyperopt）
• [recommand] including using .pickle to store data!!
  关于sklearn的网格搜索GridSearchCV寻找最优超参数

from sklearn.externals import joblib
# store model
joblib.dump(grid_search.best_estimator_, '/home/yummy/sklearn_model/xxx.pickle')
# load model
model = joblib.load('/home/yummy/sklearn_modle/xxx.pickle')
y_pred = model.predict(X_test)

• using n_jobs = -1 for parallel computation, so fast!!

elm_gcv = GridSearchCV(elm_model, elm_param_grid, cv=5, scoring='neg_mean_squared_error', verbose = 1, n_jobs = -1)

Example

• example of DecisionTree
  交叉驗證（Cross Validation）與網格搜尋（Grid Search）的原理及實證分析
• sklearn中的GridSearchCV()！python中模型里各种参数取值有它方便多了