Hi,
I have some questions on the consistency of implementation of dropping and paper.
-
When you take the sum, you did not use absolute values as written in the paper.
|
di = wi.sum(0).sort()[1][self.count:self.count + delta] |
-
You drop during the stage, not when the stage finishes, as written in the paper.
|
if not self._reach_stage(stage): |
Am I wrong or would you explain about it ? Thank you.
Hi,
I have some questions on the consistency of implementation of dropping and paper.
When you take the sum, you did not use absolute values as written in the paper.
CondenseNet/layers.py
Line 86 in 833a91d
You drop during the stage, not when the stage finishes, as written in the paper.
CondenseNet/layers.py
Line 62 in 833a91d
Am I wrong or would you explain about it ? Thank you.