Sklearn相关

1、添加权重

clf = RandomForestClassifier(n_estimators=10,class_weight ={0:0.81,1:0.19})

2、输出

pred = clf.predict_proba(test)#为概率

pred = clf.predict(test)#为结果

 3、结果集分布

group_df = train.标签.value_counts().reset_index()
k = group_df['标签'].sum()
print((group_df.标签/k))