-
Notifications
You must be signed in to change notification settings - Fork 2.5k
please set 'STRIDE_IN_1X1' to be 'False' when backbone uses GN #465
Conversation
…d use 'BaseStem'&'Bottleneck' to simply codes
…ount of indentation and code duplication
…, and add dilation into ResNetHead
Thank you for your pull request and welcome to our community. We require contributors to sign our Contributor License Agreement, and we don't seem to have you on file. In order for us to review and merge your code, please sign up at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need the corporate CLA signed. If you have received this in error or have any questions, please contact us at cla@fb.com. Thanks! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks!
Can I merge this now or do you want me to wait (maybe for some results to finish computing)? |
hi @fmassa |
Let's do it like the following: once the trainings are over, you can send another PR updating maybe the |
ok. |
Hi fmassa, I have a question here: What does the STRIDE_IN_1X1 mean? If I use ResNeXt101-FPN, should I set this option to True? Thank you for taking the time ! @fmassa |
…ookresearch#465) * make pixel indexes 0-based for bounding box in pascal voc dataset * replacing all instances of torch.distributed.deprecated with torch.distributed * replacing all instances of torch.distributed.deprecated with torch.distributed * add GroupNorm * add GroupNorm -- sort out yaml files * use torch.nn.GroupNorm instead, replace 'use_gn' with 'conv_block' and use 'BaseStem'&'Bottleneck' to simply codes * modification on 'group_norm' and 'conv_with_kaiming_uniform' function * modification on yaml files in configs/gn_baselines/ and reduce the amount of indentation and code duplication * use 'kaiming_uniform' to initialize resnet, disable gn after fc layer, and add dilation into ResNetHead * agnostic-regression for bbox * please set 'STRIDE_IN_1X1' to be 'False' when backbone use GN
to fix the performance gap compared to Detectron.