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
Currently, the torch_tensor_builder and torch_tensor_resolver are cast to numpy builder and resolver directly, and bfloat16 is cast to float16 which caused the loss of accuracy.
The text was updated successfully, but these errors were encountered:
Describe your problem
Currently, the torch_tensor_builder and torch_tensor_resolver are cast to numpy builder and resolver directly, and bfloat16 is cast to float16 which caused the loss of accuracy.
The text was updated successfully, but these errors were encountered: