Hi, thanks for sharing the code. I am implementing the CNN part. I think the block_idx in forward function should be moded by 2 when CNN case as you used only two softmax. Could you check it? Thanks.
|
for block_idx in range(2*(self.args.num_blocks - 1) + 1): |
|
logits, hidden = self.forward(inputs, |
|
hidden, |
|
block_idx, |
|
is_embed=(block_idx == 0)) |
Hi, thanks for sharing the code. I am implementing the CNN part. I think the
block_idxinforwardfunction should be moded by 2 when CNN case as you used only two softmax. Could you check it? Thanks.ENAS-pytorch/models/controller.py
Lines 173 to 177 in 25c4a89