Tensorflow2.0报错：ProfilerNotRunningError: Cannot stop profiling. No profiler is running.(修改后别忘了重启内核或关掉重启)

# 把这个路径用包装一下
logdir = os.path.join("cnn_selu_callbacks")

print(logdir)
if not os.path.exists(logdir):
    os.mkdir(logdir)
output_model_file = os.path.join(logdir,
                                 "fashion_mnist_model.h5")

callbacks = [
    keras.callbacks.TensorBoard(log_dir=logdir),
    keras.callbacks.ModelCheckpoint(output_model_file,
                                    save_best_only = True),
    keras.callbacks.EarlyStopping(patience=5, min_delta=1e-3),
]
history = model.fit(x_train_scaled, y_train, epochs=10,
                    validation_data=(x_valid_scaled, y_valid),
                    callbacks = callbacks)
————————————————
版权声明：本文为CSDN博主「随海亮」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/weixin_41010198/article/details/102815483

原文地址：https://www.cnblogs.com/qqhfeng/p/12154514.html