Apache MXNet Forum
Implementation of weighted softmax by extending mx.autograd.Function fails
ThomasDelteil
September 2, 2019, 12:42am
2
See my other answer to your post: I would recommend using a custom op:
show post in topic
Related Topics
Topic
Replies
Views
Activity
About stale gradient
Gluon
17
3199
October 19, 2020
Understanding Autograd.backward() with custom parameters for specific layers
Discussion
3
780
September 17, 2019
Cannot bind model with custom loss function
Discussion
1
1380
January 24, 2018
Multiple losses
Gluon
7
3650
June 5, 2018
WGAN-gp: can't compute gradient penalty with gluon?
Gluon
0
407
October 15, 2020