import csv
with open ('news_train.txt', 'r', encoding='utf-8') as news_train:
    train=list(csv.reader(news_train, delimiter='\t'))

trainCat=[None]*60000
trainNews=[None]*60000
i=0
for element in train:
    trainCat[i]=train[i][0]
    trainNews[i]=train[i][1]+' '+train[i][2]
    i=i+1

with open ('news_test.txt', 'r', encoding='utf-8') as news_test:
    test=list(csv.reader(news_test, delimiter='\t'))

testNews=[None]*15000
i=0
for element in test:
    testNews[i]=test[i][0]+' '+test[i][1]
    i=i+1

from sklearn.svm import LinearSVC
from sklearn.pipeline import Pipeline
from sklearn.multiclass import OneVsRestClassifier
from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer

predictor=Pipeline([
        ('vectorizer', CountVectorizer()),
        ('tfidf', TfidfTransformer()),
        ('clf', OneVsRestClassifier(LinearSVC())),
    ])
predictor.fit(trainNews, trainCat)
result=predictor.predict(testNews)

with open ('result.txt', 'w', encoding='utf-8') as predResult:
    for element in result:
        predResult.write(element+'\n')