Colab+Keras+TensorBoard FailedPreconditionError

Question

我正在尝试运行一个简单的 Keras 脚本并使用 Google Colab 和 TensorBoard。这是我的代码：

import tensorflow as tf
import tensorflow.keras as keras
from tensorflow.keras.datasets import cifar10
from tensorflow.keras.applications.mobilenet import MobileNet
from tensorboardcolab import TensorBoardColab, TensorBoardColabCallback

# Settings
num_classes = 10
batch_size = 16
epochs = 1

# Data setup
(x_train, y_train), (x_test, y_test) = cifar10.load_data()
x_train = x_train.astype('float32') / 255
x_test = x_test.astype('float32') / 255
y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)

# Select model
model = MobileNet(weights=None, input_shape=x_train.shape[1:], classes=num_classes)

# Select loss, optimizer, metric
model.compile(loss='categorical_crossentropy',
                            optimizer=tf.train.AdamOptimizer(0.001),
                            metrics=['accuracy'])    
# Train
tbc=TensorBoardColab()
model.fit(x_train, y_train,
                    batch_size=batch_size,
                    epochs=epochs,
                    verbose=1,
                    validation_data=(x_test, y_test), 
                    callbacks=[TensorBoardColabCallback(tbc)])

这是我看到的将 TensorBoard 与 Colab 结合使用的建议，参考此处：

但是，在添加回调时出现错误：

FailedPreconditionError: Error while reading resource variable conv_dw_8_2/depthwise_kernel from Container: localhost. This could mean that the variable was uninitialized. Not found: Resource localhost/conv_dw_8_2/depthwise_kernel/N10tensorflow3VarE does not exist. [[Node: conv_dw_8_2/depthwise/ReadVariableOp = ReadVariableOpdtype=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]] [[Node: loss_2/mul/_147 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge_6752_loss_2/mul", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]

有人知道我做错了什么吗？这似乎是运行 Colab 上 TensorBoard 的一种非常有用的方法，如果我能让它工作的话。

Answer 1

这是Keras版本冲突造成的。 Tensorboardcolab uses the full keras library while you import the tf.keras 实施 Keras API。因此，当您拟合模型时，您最终会使用两个不同版本的 keras。

您有几个选择：

使用 Keras 库并更改导入

import tensorflow as tf
import keras
from keras.datasets import cifar10
from keras.applications.mobilenet import MobileNet
from tensorboardcolab import TensorBoardColab, TensorBoardColabCallback

虽然代码在这些更改下运行良好，但您可以考虑使用 Keras's version of the Adam optimizer，因此您不再需要显式导入 tensorflow。

model.compile(loss='categorical_crossentropy', 
                    optimizer=keras.optimizers.Adam(lr=0.001), 
                    metrics=['accuracy'])`

使用 tf.keras 并修补 TensorBoardColab

如果您修补 callbacks.py and core.py 并修复那里的导入，您的代码运行良好：

<s>从 keras.callbacks 导入 TensorBoard</s> 来自 tensorflow.keras.callbacks 导入 TensorBoard

您也可以在我进行这些更改的地方使用 this fork。

Colab+Keras+TensorBoard FailedPreconditionError