anirudh2290 commented on a change in pull request #15118: Conversion from FP32
model to Mixed Precision model
URL: https://github.com/apache/incubator-mxnet/pull/15118#discussion_r293171606
##########
File path: python/mxnet/gluon/parameter.py
##########
@@ -919,23 +919,49 @@ def load(self, filename, ctx=None, allow_missing=False,
assert name.startswith(restore_prefix), \
"restore_prefix is '%s' but Parameters name '%s' does not
start " \
"with '%s'"%(restore_prefix, name, restore_prefix)
- lprefix = len(restore_prefix)
ndarray_load = ndarray.load(filename)
+ self.load_dict(ndarray_load, ctx, allow_missing,
+ ignore_extra, restore_prefix, filename, cast_dtype)
+
+ def load_dict(self, param_dict, ctx=None, allow_missing=False,
Review comment:
yes it is related to the PR, it is used in the convert_hybrid_block
----------------------------------------------------------------
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
