yep!
Sorry, i'm beginner.
i try to use
y = np.ones([2*batch_size, 1])
y[batch_size:, :] = 0
instead of y = [1] batch_size + [0] batch_size
Then d_loss = d.train_on_batch(x, y) is OK!
z = np.ones([batch_size, 1])
When i calling d_on_g_loss = d_on_g.train_on_batch(image_blur_batch,z)
I found Warning that says /path/to/keras/engine/training.py:973: UserWarning: Discrepancy between trainable weights and collected trainable weights, did you set model.trainable without calling model.compile after ? 'Discrepancy between trainable weights and collected trainable'
@leftthomas
yep! Sorry, i'm beginner. i try to use y = np.ones([2*batch_size, 1]) y[batch_size:, :] = 0
instead of y = [1] batch_size + [0] batch_size
Then d_loss = d.train_on_batch(x, y) is OK!
z = np.ones([batch_size, 1]) When i calling d_on_g_loss = d_on_g.train_on_batch(image_blur_batch,z)
I found Warning that says /path/to/keras/engine/training.py:973: UserWarning: Discrepancy between trainable weights and collected trainable weights, did you set model.trainable without calling model.compile after ? 'Discrepancy between trainable weights and collected trainable'
Thank you for your reply.