-
Notifications
You must be signed in to change notification settings - Fork 433
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Feature] Add Maskfeat Support #485
Conversation
Codecov ReportBase: 72.48% // Head: 73.26% // Increases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## dev #485 +/- ##
==========================================
+ Coverage 72.48% 73.26% +0.78%
==========================================
Files 121 125 +4
Lines 4765 4994 +229
Branches 763 788 +25
==========================================
+ Hits 3454 3659 +205
- Misses 1151 1167 +16
- Partials 160 168 +8
Flags with carried forward coverage won't be shown. Click here to find out more.
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
drop_path_rate=0, | ||
), | ||
head=dict(type='MaskFeatPretrainHead', hog_dim=108), | ||
hog_para=dict(nbins=9, pool=8, gaussian_window=16)) |
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.
please add some comments for hog related parameters, to help the users.
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.
As inline comments
configs/benchmarks/classification/imagenet/maskfeat_vit-base-p16_ft-8xb512-coslr-100e_in1k.py
Outdated
Show resolved
Hide resolved
|
||
|
||
@PIPELINES.register_module() | ||
class MaskfeatMaskGenerator(object): |
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.
How about using MaskFeatureMaskGenerator
? @fangyixiao18
class MaskfeatMaskGenerator(object): | |
class MaskFeatMaskGenerator(object): |
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.
I think it is better to use MaskFeatMaskGenerator
,consisting with algorithm name in our implementation
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.
Update the name and this pr could be merged.
* [Feature]: Add MaskfeatMaskGenerator Pipeline * [Feature]: Add HogLayerC for MaskFeat * [Feature]: Add Backbone of MaskFeat * [Feature]: Add Head of MaskFeat * [Feature]: Add Algorithms of MaskFeat * [Feature]: Add Config of MaskFeat * [Doc] Update Readme of MaskFeat * [Fix] fix ut and hog_layer. * [fix] Add and correct docstring * [Fix] Refine the docstring of MaskFeat * [fix] fix value of trunc_normal_ * [fix] rename the finetune config of maskfeat * [fix] rename the fine-tuning config of maskfeat * [fix] rename the fine-tuning config of maskfeat * [fix] add new paramwise_options in fine-tuning config * [fix] update the top-1 accuary of maskfeat * [fix] update the top-1 accuary of maskfeat in model_zoo * [fix] rename MaskfeatMaskGenerator
* [Feature]: Add MaskfeatMaskGenerator Pipeline * [Feature]: Add HogLayerC for MaskFeat * [Feature]: Add Backbone of MaskFeat * [Feature]: Add Head of MaskFeat * [Feature]: Add Algorithms of MaskFeat * [Feature]: Add Config of MaskFeat * [Doc] Update Readme of MaskFeat * [Fix] fix ut and hog_layer. * [fix] Add and correct docstring * [Fix] Refine the docstring of MaskFeat * [fix] fix value of trunc_normal_ * [fix] rename the finetune config of maskfeat * [fix] rename the fine-tuning config of maskfeat * [fix] rename the fine-tuning config of maskfeat * [fix] add new paramwise_options in fine-tuning config * [fix] update the top-1 accuary of maskfeat * [fix] update the top-1 accuary of maskfeat in model_zoo * [fix] rename MaskfeatMaskGenerator
* [Feature]: Add MaskfeatMaskGenerator Pipeline * [Feature]: Add HogLayerC for MaskFeat * [Feature]: Add Backbone of MaskFeat * [Feature]: Add Head of MaskFeat * [Feature]: Add Algorithms of MaskFeat * [Feature]: Add Config of MaskFeat * [Doc] Update Readme of MaskFeat * [Fix] fix ut and hog_layer. * [fix] Add and correct docstring * [Fix] Refine the docstring of MaskFeat * [fix] fix value of trunc_normal_ * [fix] rename the finetune config of maskfeat * [fix] rename the fine-tuning config of maskfeat * [fix] rename the fine-tuning config of maskfeat * [fix] add new paramwise_options in fine-tuning config * [fix] update the top-1 accuary of maskfeat * [fix] update the top-1 accuary of maskfeat in model_zoo * [fix] rename MaskfeatMaskGenerator
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.
Motivation
Add maskfeat support.
Modification
Add MaskGenerator.
Add backbone of maskfeat.
Add head of maskfeat.
Add algorithms of maskfeat.
Add config of maskfeat.
Add Config of MaskFeat.
Update Readme of MaskFeat.
BC-breaking (Optional)
Does the modification introduce changes that break the backward compatibility of the downstream repositories?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.
Use cases (Optional)
If this PR introduces a new feature, it is better to list some use cases here and update the documentation.
Checklist
Before PR:
After PR: