You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.
Currently when storage fallback happens, the log message is printed in the console, but it's very hard for the user to figure out when the fallback happens in the code even if they use NaiveEngine for debugging. We should throw an exception when storage fallback happens to help the user debug it. https://github.com/apache/incubator-mxnet/blob/master/src/operator/operator_common.h#L526
missing storage fallback message/exception during ndarray copy
cast_storage may happen when the source and destination ndarray are of different storage types. Currently it's converted in silence and the user is not aware.
sparse vector with shape=(10)
#8817
Still under discussion. We should either fix it or at least not let VM crash.
throw exception on storage fallback
Currently when storage fallback happens, the log message is printed in the console, but it's very hard for the user to figure out when the fallback happens in the code even if they use NaiveEngine for debugging. We should throw an exception when storage fallback happens to help the user debug it.
https://github.com/apache/incubator-mxnet/blob/master/src/operator/operator_common.h#L526
missing storage fallback message/exception during ndarray copy
cast_storage may happen when the source and destination ndarray are of different storage types. Currently it's converted in silence and the user is not aware.
sparse vector with shape=(10)
#8817
Still under discussion. We should either fix it or at least not let VM crash.
distributed training tutorial for sparse
Maybe do this after kvstore=device is done.
Testing
#8709
#8542
#8980
The text was updated successfully, but these errors were encountered: