eric-haibin-lin commented on a change in pull request #16893: Multi-tensor LAMB
URL: https://github.com/apache/incubator-mxnet/pull/16893#discussion_r352875009
##########
File path: python/mxnet/ndarray/contrib.py
##########
@@ -605,3 +605,27 @@ def multi_mp_adamw_update(weights, grads, mean, var,
weights32, rescale_grad, lr
etas=etas,
name=name,
**kwargs)
+
+def multi_lamb_update(weights, grads, mean, var, temp_g, step_count,
+ out=None, name=None, num_tensors=0, **kwargs):
Review comment:
Could you add some documentation on the input types for these two functions?
Otherwise it shows nothing on the API website
----------------------------------------------------------------
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]
With regards,
Apache Git Services
