Hey. I am attempting to run the code (sample.py) to see how it works. However, I made the changes in TF 2 that I should, but I now get an error "TypeError: 'str' object is not callable" I don't understand why. I would appreciate any help you can provide.
This is where the error occurs:
net_h0 = layers.Dense(units=gf_dim * 4 * s8 * s8, activation=tf.identity,
kernel_initializer=w_init, name='g/h0/lin')(net_in)
in vae_coding.py in the decoder.
Same error i have when i run vae_train_anime.py in this line (encoder):
net_h0 = Conv2d(ef_dim, (5, 5), (2, 2), act=None,
padding='SAME', W_init=w_init, name='en/h0/conv2d')(net_in)
I think it's related to the generator, in function build_graph that calls this function ( in this lines:
with tf.compat.v1.variable_scope('generator'):
self.generated = self._generator(self.candid, self._latent_dim)), but I'm not sure what exactly it is.
Thank you in advance
Hey. I am attempting to run the code (sample.py) to see how it works. However, I made the changes in TF 2 that I should, but I now get an error "TypeError: 'str' object is not callable" I don't understand why. I would appreciate any help you can provide.
This is where the error occurs:
net_h0 = layers.Dense(units=gf_dim * 4 * s8 * s8, activation=tf.identity,
kernel_initializer=w_init, name='g/h0/lin')(net_in)
in vae_coding.py in the decoder.
Same error i have when i run vae_train_anime.py in this line (encoder):
net_h0 = Conv2d(ef_dim, (5, 5), (2, 2), act=None,
padding='SAME', W_init=w_init, name='en/h0/conv2d')(net_in)
I think it's related to the generator, in function build_graph that calls this function ( in this lines:
with tf.compat.v1.variable_scope('generator'):
self.generated = self._generator(self.candid, self._latent_dim)), but I'm not sure what exactly it is.
Thank you in advance