How to fix certain parameters during gluon training?


Setattr(“grad_req”, “null”) works only for setting gradient of a block of parameters. I want to set “grad_req”, “null” for a particular parameter of the network.


I think you can reference to this post


Is there a way to fix running_mean and running_var and not compute it based on the current batch?


Maybe you can set use_global_stats=True ?