西瓜书--线性回归 logistic 01

import pandas as pd

负责数据的读写和初步处理

from sklearn.linear_model import LogisticRegression

调用线性回归模型logistic

from sklearn.feature_extraction_text import CountVectorize

调用文本特征提取中的统计词频功能

df_train=pd.read_csv(‘./train_set.csv’)

df_test=pd.read_csv(‘./test_set.csv’)

df_train.drop(columns=[‘Id’ ,’article’],inplace=True)

df_test.drop(columns=[‘article ’],inplace=True)

vectorize=CountVectorize(ngram_rang=(1,2),min_df=3,max_df=0.9,max_features=100000)

vectorize.fit(df_train[‘word_seg’])

x_train=vectorize.transform(df_train[‘word_seg’])

y_train=df_train[‘class’]-1

x_test=vectorize.transform(df_test[‘word_seg’])

lg=LogisticRegression(C=4,dual=True)

lg.fit(x_train,y_train)

y_test=lg.predict(x_test)

df_test[‘class’]=y_test.tolist()

df_test[‘class’]+=1

df_result=df_test.loc[:,[‘Id ’,’class’]]

df_result.to_csv(‘result.csv’,index=False)