df1 = pd.read_csv('drive/MyDrive/Bert_learn/train.csv',encoding='utf8')
Toxicity comment classification by using BERT and BiLSTM model train more epochs to get good accuracy
hist = model.fit([X_train_id, X_train_mask], y_train, validation_data=([X_test_id, X_test_mask], y_test), epochs=1, batch_size=64)