Skip to content
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] Release new fp16-dcn models ( mask-rcnn DCN/DCNv2) #5201

Merged
merged 3 commits into from
May 23, 2021

Conversation

AronLin
Copy link
Contributor

@AronLin AronLin commented May 21, 2021

Motivation

Release mask_rcnn_r50_fpn_fp16_dconv_c3-c5_1x_coco and mask_rcnn_r50_fpn_fp16_mdconv_c3-c5_1x_coco,
including config files, models and log files.

it is based on open-mmlab/mmcv#1029.

Modification

Add two config files.
Release models.
Modify the readme file.

BC-breaking (Optional)

If pytorch >= 1.6.0 with that the mmcv version is larger than 1.3.2 and before open-mmlab/mmcv#1029 merged, there will be a type mismatch bug in fp16 training.

@CLAassistant
Copy link

CLAassistant commented May 21, 2021

CLA assistant check
All committers have signed the CLA.

@AronLin AronLin requested a review from ZwwWayne May 21, 2021 13:53
@AronLin AronLin changed the title [Feature] Release new fp16 models ( mask-rcnn DCN/DCNv2) [Feature] Release new fp16-dcn models ( mask-rcnn DCN/DCNv2) May 21, 2021
@codecov
Copy link

codecov bot commented May 21, 2021

Codecov Report

Merging #5201 (35febe6) into master (3d91b8b) will decrease coverage by 0.86%.
The diff coverage is n/a.

❗ Current head 35febe6 differs from pull request most recent head 818e9a8. Consider uploading reports for the commit 818e9a8 to get more accurate results
Impacted file tree graph

@@            Coverage Diff             @@
##           master    #5201      +/-   ##
==========================================
- Coverage   65.18%   64.32%   -0.87%     
==========================================
  Files         272      272              
  Lines       21078    20933     -145     
  Branches     3531     3531              
==========================================
- Hits        13740    13465     -275     
- Misses       6570     6687     +117     
- Partials      768      781      +13     
Flag Coverage Δ
unittests 64.31% <ø> (-0.84%) ⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files Coverage Δ
mmdet/models/necks/yolo_neck.py 62.00% <0.00%> (-38.00%) ⬇️
mmdet/models/dense_heads/yolo_head.py 27.90% <0.00%> (-23.25%) ⬇️
mmdet/core/export/onnx_helper.py 13.18% <0.00%> (-19.08%) ⬇️
mmdet/models/dense_heads/fcos_head.py 67.84% <0.00%> (-7.05%) ⬇️
mmdet/models/dense_heads/rpn_test_mixin.py 77.41% <0.00%> (-6.46%) ⬇️
mmdet/core/bbox/coder/delta_xywh_bbox_coder.py 94.89% <0.00%> (-5.11%) ⬇️
mmdet/models/utils/normed_predictor.py 88.09% <0.00%> (-4.77%) ⬇️
mmdet/core/bbox/transforms.py 75.89% <0.00%> (-4.47%) ⬇️
mmdet/models/dense_heads/anchor_head.py 83.62% <0.00%> (-3.10%) ⬇️
mmdet/models/detectors/base.py 52.54% <0.00%> (-2.66%) ⬇️
... and 54 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 3d91b8b...818e9a8. Read the comment docs.

@ZwwWayne ZwwWayne merged commit 0247995 into open-mmlab:master May 23, 2021
@AronLin AronLin deleted the fp16 branch May 26, 2021 14:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants