Mengmeng bai

your GPU memory is small

line 128 and 162 in the model.py with tf.name_scope('source_train_op') modifies to with tf.variable_scope('source_train_op',reuse=False): with tf.name_scope('target_train_op') modifes to with tf.variable_scope('target_train_op',reuse=False):

> Hi there, > > I have noticed that you have set the ratio between `l1 loss`, `perceptual loss` and `adversarial loss` as `100: 10: 1`, which is not the...

> Interesting, I would expect positional encoding (possibly a different encoding than a simple linear mesh) would have helped. > > So, this suggests a few possible outcomes (1x1 filter/conv...