警告：tensorflow：在 TensorFlow 2.0 中，“TensorBoard”回调将忽略“write_grads”

Question

我使用以下代码行使用张量板来可视化 ANN 模型的梯度

  tensorboard_callback = tf.compat.v1.keras.callbacks.TensorBoard(log_dir='./Graph', histogram_freq=1, write_graph = True, write_grads =True, write_images = False)

tensorboard_callback .set_model(model)


%tensorboard --logdir ./Graph

我收到一条警告消息，内容为“警告：tensorflow：回调write_grads将在 TensorFlow 2.0 中被忽略TensorBoard”。

我得到张量板输出，但没有梯度。

可能的原因是什么？

（注：我使用的是2.3.0tensorflow版本）

谢谢。

Answer 1

Write_Grads没有在 中实施TF2.x。这是仍处于开放状态的备受期待的功能请求之一。请检查此 GitHub问题作为功能请求。所以，我们只需要导入TF1.x模块并使用，write_grads如下面的代码所示。

# Load the TensorBoard notebook extension
%load_ext tensorboard

import tensorflow as tf
import datetime

# Clear any logs from previous runs
!rm -rf ./logs/ 

# Disable V2 behavior
tf.compat.v1.disable_v2_behavior()

mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()

x_train, x_test = x_train / 255.0, x_test / 255.0


def create_model():

  return tf.keras.models.Sequential([

    tf.keras.layers.Flatten(input_shape=(28, 28)),

    tf.keras.layers.Dense(512, activation='relu'),

    tf.keras.layers.Dropout(0.2),

    tf.keras.layers.Dense(10, activation='softmax')

  ])

 

model = create_model()

model.compile(optimizer='adam',

              loss='sparse_categorical_crossentropy',

              metrics=['accuracy'])


log_dir = "logs/fit/" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S")

tensorboard_callback = tf.compat.v1.keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1, write_grads =True)

model.fit(x=x_train, y=y_train, epochs=1, validation_data=(x_test, y_test), callbacks=[tensorboard_callback]) 

%tensorboard --logdir logs/fit

输出：

Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz
11493376/11490434 [==============================] - 0s 0us/step

Train on 60000 samples, validate on 10000 samples
WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/engine/training_v1.py:2048: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.
Instructions for updating:
This property should not be used in TensorFlow 2.0, as updates are applied automatically.
   32/60000 [..............................] - ETA: 0s - loss: 2.3311 - acc: 0.0312WARNING:tensorflow:Callbacks method `on_train_batch_end` is slow compared to the batch time (batch time: 0.0055s vs `on_train_batch_end` time: 0.0235s). Check your callbacks.
60000/60000 [==============================] - 17s 288us/sample - loss: 0.2187 - acc: 0.9349 - val_loss: 0.1012 - val_acc: 0.9690
<tensorflow.python.keras.callbacks.History at 0x7f7ebd1d3d30>