ThomasDelteil commented on issue #8066: How to freeze mean and variance parameters of batchnorm layers of resnet-50? When using mx.mod.Module(fixed_param_names), those parameters belongs to auxiliary parameters and not in symbol.list_arguments() URL: https://github.com/apache/incubator-mxnet/issues/8066#issuecomment-402567054 @jay1204 does @solin319 reply answers your question? If not can you provide more context as to what you are trying to achieve? Are you trying to perform batch normalization with fixed values during training? @eric-haibin-lin can you please close the issue? Thanks! @jay1204 if you would like to follow up, please create a post in https://discuss.mxnet.io and reference this github issue, thanks!
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on 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
