TFIDF+LR.py

import numpy as np
import pandas as pd

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_val_score
from scipy.sparse import hstack

class_names = ['toxic', 'severe_toxic', 'obscene', 'threat', 'insult', 'identity_hate']

train = pd.read_csv('../input/train.csv').fillna(' ')
test = pd.read_csv('../input/test.csv').fillna(' ')

train_text = train['comment_text']
test_text = test['comment_text']
all_text = pd.concat([train_text, test_text])

word_vectorizer = TfidfVectorizer(
    sublinear_tf=True,
    strip_accents='unicode',
    analyzer='word',
    token_pattern=r'\w{1,}',
    stop_words='english',
    ngram_range=(1, 1),
    max_features=10000)
word_vectorizer.fit(all_text)
train_word_features = word_vectorizer.transform(train_text)
test_word_features = word_vectorizer.transform(test_text)

char_vectorizer = TfidfVectorizer(
    sublinear_tf=True,
    strip_accents='unicode',
    analyzer='char',
    stop_words='english',
    ngram_range=(2, 6),
    max_features=50000)
char_vectorizer.fit(all_text)
train_char_features = char_vectorizer.transform(train_text)
test_char_features = char_vectorizer.transform(test_text)

train_features = hstack([train_char_features, train_word_features])
test_features = hstack([test_char_features, test_word_features])

scores = []
submission = pd.DataFrame.from_dict({'id': test['id']})
for class_name in class_names:
    train_target = train[class_name]
    classifier = LogisticRegression(C=0.1, solver='sag')

    cv_score = np.mean(cross_val_score(classifier, train_features, train_target, cv=3, scoring='roc_auc'))
    scores.append(cv_score)
    print('CV score for class {} is {}'.format(class_name, cv_score))

    classifier.fit(train_features, train_target)
    submission[class_name] = classifier.predict_proba(test_features)[:, 1]

print('Total CV score is {}'.format(np.mean(scores)))

from sklearn.metrics import roc_auc_score
test_label=pd.read_csv('../input/test_labels.csv')
auc_sum = 0
for class_ in class_names:
    sub_test_label=test_label[test_label.id.isin(test_label[test_label[class_]==-1].id.tolist())==False]
    sub_submission=submission[submission.id.isin(test_label[test_label[class_]==-1].id.tolist())==False]
    auc_sum += roc_auc_score(sub_test_label[class_],sub_submission[class_])
print("test_average_auc_score:",auc_sum/len(class_names))
#备注：如果test_label=-1,该样本的不计入auc的计算中。