InstanceNorm{1,2,3}d (#28790)

Summary:
Hi yf225,

I have a few doubts related to implementation:
1) What tests do I have to write?
2) What does _load_state_from_dict does?
3) Do I need to override reset() function as I can not see it's utility?
4) InstanceNormOptions could be removed with BatchNormOptions, but I find that
`track_running_status` is not defined instead `stateful` is defined.

InstanceNorm{1,2,3}d https://github.com/pytorch/pytorch/issues/25883
Pull Request resolved: https://github.com/pytorch/pytorch/pull/28790

Differential Revision: D18588666

Pulled By: yf225

fbshipit-source-id: bb9b81f01f62c3fc8765fa0ba0716768087ee155
This commit is contained in:
Divyansh Singhvi
2019-11-19 16:53:37 -08:00
committed by Facebook Github Bot
parent 8e3486de81
commit ec52d911bd
15 changed files with 880 additions and 115 deletions

View File

@ -75,9 +75,9 @@ torch.nn.BatchNorm2d|Yes|No
torch.nn.BatchNorm3d|Yes|No
torch.nn.GroupNorm|Yes|No
torch.nn.SyncBatchNorm|No|No
torch.nn.InstanceNorm1d|No|No
torch.nn.InstanceNorm2d|No|No
torch.nn.InstanceNorm3d|No|No
torch.nn.InstanceNorm1d|Yes|No
torch.nn.InstanceNorm2d|Yes|No
torch.nn.InstanceNorm3d|Yes|No
torch.nn.LayerNorm|Yes|No
torch.nn.LocalResponseNorm|Yes|No
torch.nn.CrossMapLRN2d|Yes|No