Light-- commented on issue #8623:
URL:
https://github.com/apache/incubator-mxnet/issues/8623#issuecomment-675902054
> right, I don't think it's exposed in the frontend. lr_mult should be set
on a variable such as the weight of fullyconnected.
ok, since i can't find any official example for doing this, now i do it by
myself:
1, is the following code right?
```
_weight = mx.symbol.Variable("fc1_weight", lr_mult=10, )
fc1 = mx.sym.FullyConnected(data=fc0, num_hidden=num_classes, name='fc1')
```
2, do i need to set the lr_mut for optimizer at the same time to make it
work? or just set one of them is enough? [some post said it also need to set
lr_mult for the optimizer](https://discuss.gluon.ai/t/topic/6002/12)
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
