Thanks for the nice code, but I find some problems here.
When I'm trying to run train.py, it returns some error as
ValueError: Variable discriminator/layer_0/batch_norm/discriminator_1/layer_0/batch_norm/moments/normalize/mean/ExponentialMovingAverage/ does not exist, or was not created with tf.get_variable(). Did you mean to set reuse=None in VarScope?
Do you have any ideas how to solve for it?
Also, there may be some misunderstandings in you implementation procedure. In the README file, you suggest first run python3 infogan/__init__.py --infogan to run infogan. But there would be some importing problems. Of course you can directly run train.py in the home directory, but it would be better if you can make it clear how to run the init file first.
Hi there,
Thanks for the nice code, but I find some problems here.
When I'm trying to run train.py, it returns some error as
ValueError: Variable discriminator/layer_0/batch_norm/discriminator_1/layer_0/batch_norm/moments/normalize/mean/ExponentialMovingAverage/ does not exist, or was not created with tf.get_variable(). Did you mean to set reuse=None in VarScope?Do you have any ideas how to solve for it?
Also, there may be some misunderstandings in you implementation procedure. In the README file, you suggest first run
python3 infogan/__init__.py --infoganto run infogan. But there would be some importing problems. Of course you can directly runtrain.pyin the home directory, but it would be better if you can make it clear how to run the init file first.Thanks in advance