90. model.evaluate 和 model.predict 的区别

[CHH3213]

https://blog.csdn.net/DoReAGON/article/details/88552348

-model.evaluate

输入数据和标签,输出损失和精确度.

    # 评估模型,不输出预测结果
    loss,accuracy = model.evaluate(X_test,Y_test)
    print('\ntest loss',loss)
    print('accuracy',accuracy)

• model.predict 输入测试数据,输出预测结果 (通常用在需要得到预测结果的时候)

  
  #模型预测,输入测试集,输出预测结果
  y_pred = model.predict(X_test,batch_size = 1)