from sklearn.linear_model import LogisticRegressionCV
X_train_stem_u = vectorizer_stem_u.fit_transform(X_train)

X_prepeared_data_train = X_train_stem_u
selected_vectorizer = vectorizer_stem_u

clf =  LogisticRegressionCV(cv=5,max_iter=800).fit(X_prepeared_data_train, y_train)#solver='liblinear',penalty='l1',

X_prepeared_data_test = selected_vectorizer.transform(X_test)



predicted = clf.predict(X_prepeared_data_test)
print(np.mean(predicted == y_test))

import eli5

eli5.show_weights(clf, vec=selected_vectorizer, top=1500,
                  target_names=["group1","group2"])

+7.323  peter
+6.143  ing
+6.033  enabl
+5.918  anand
+5.893  jose