@yuewu001 1. Trainer is now fixed and uses the same logic as in Module. [Here in Trainer](- https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/gluon/trainer.py#L188) is where you create the KVStore using [model._create_kvstore](https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/model.py#L77). However, please note that, it is not recommended to use KVStore when grad is sparse, hence, it will be set to false when grad is sparse. 2. To save optimizer states, you can use trainer.save_states(file_name). https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/gluon/trainer.py#L376
Resolving the issue. Please reopen, if your still have questions/issues. [ Full content available at: https://github.com/apache/incubator-mxnet/issues/9557 ] This message was relayed via gitbox.apache.org for [email protected]
