convolutional_network_raw.ipynb doens't work for python3
seilgu opened this issue · comments
seilgu commented
trainable_variables = weights.values() + biases.values()
doesn't work.
Also if I write trainable_variables = list(weights.values()) + list(biases.values())
, I have to turn on tf.enable_eager_execution()
, but the training result is wrong, accuracy is around 0.1 and not improving.
Yui commented
Convert dict_values to list. It works well
trainable_variables = list(weights.values()) + list(biases.values())