tensorflow2.x 使用callback方法停止模型训练(mnist手写数字)

最新推荐文章于 2024-04-23 01:33:40 发布

怅然一梦，烟火星辰

最新推荐文章于 2024-04-23 01:33:40 发布

阅读量684

点赞数

文章标签： tensorflow 深度学习神经网络

本文链接：https://blog.csdn.net/weixin_46016713/article/details/120788385

版权

import tensorflow as tf
from tensorflow.keras import layers

# 定义一个myCallback类，继承了tensorflow中自带的Callback类
class myCallback(tf.keras.callbacks.Callback):
    def on_epoch_end(self, epoch, logs=None):
        if(logs.get('accuracy') > 0.96):
            print("\n 已经到达 96% 的训练精度！")
            self.model.stop_training = True

# 实例化一个myCallback 对象 callbacks
callback = myCallback()
mnist = tf.keras.datasets.mnist

(training_images, training_labels), (test_images, test_labels) = mnist.load_data()

training_images = training_images / 255.0
test_images = test_images / 255.0


model = tf.keras.models.Sequential([layers.Flatten(),
                                    layers.Dense(128, activation="relu"),
                                    layers.Dense(10, activation="softmax")
                                    ])

model.compile(optimizer="adam",
              loss="sparse_categorical_crossentropy",
              metrics=['accuracy'])

model.fit(training_images, training_labels, epochs=10, callbacks=[callback])

model.evaluate(test_images, test_labels)