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

[ModelZoo] Port CSN checkpoint from VMZ #945

Merged
merged 8 commits into from
Jul 1, 2021

Conversation

dreamerlin
Copy link
Collaborator

No description provided.

@dreamerlin dreamerlin linked an issue Jun 18, 2021 that may be closed by this pull request
@dreamerlin dreamerlin changed the title Port CSN checkpoint from VMZ [ModelZoo] Port CSN checkpoint from VMZ Jun 18, 2021
@codecov
Copy link

codecov bot commented Jun 18, 2021

Codecov Report

Merging #945 (2ad5349) into master (1f9dedc) will decrease coverage by 0.06%.
The diff coverage is n/a.

❗ Current head 2ad5349 differs from pull request most recent head 9eb1048. Consider uploading reports for the commit 9eb1048 to get more accurate results
Impacted file tree graph

@@            Coverage Diff             @@
##           master     #945      +/-   ##
==========================================
- Coverage   83.77%   83.70%   -0.07%     
==========================================
  Files         132      132              
  Lines       10056     9983      -73     
  Branches     1733     1722      -11     
==========================================
- Hits         8424     8356      -68     
+ Misses       1212     1211       -1     
+ Partials      420      416       -4     
Flag Coverage Δ
unittests 83.70% <ø> (-0.07%) ⬇️

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

Impacted Files Coverage Δ
mmaction/models/backbones/resnet3d_csn.py 97.36% <ø> (ø)
mmaction/utils/decorators.py 93.33% <0.00%> (-6.67%) ⬇️
mmaction/utils/precise_bn.py 86.95% <0.00%> (-2.60%) ⬇️
mmaction/models/backbones/tanet.py 94.23% <0.00%> (-1.93%) ⬇️
mmaction/core/evaluation/ava_utils.py 81.74% <0.00%> (-0.66%) ⬇️
.../evaluation/ava_evaluation/per_image_evaluation.py 82.85% <0.00%> (-0.25%) ⬇️
mmaction/datasets/pipelines/augmentations.py 92.75% <0.00%> (-0.23%) ⬇️
...tion/ava_evaluation/object_detection_evaluation.py 70.65% <0.00%> (-0.18%) ⬇️
mmaction/models/backbones/resnet_audio.py 81.15% <0.00%> (-0.14%) ⬇️
mmaction/models/heads/bbox_head.py 91.47% <0.00%> (ø)
... and 4 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 1f9dedc...9eb1048. Read the comment docs.

@dreamerlin
Copy link
Collaborator Author

These files are directly ported from VMZ without training, are they still needed to write into meta files?

@dreamerlin
Copy link
Collaborator Author

@kennymckormick

@kennymckormick
Copy link
Member

Please remove the fields like gpu_mem and inference_time for now (since they are not systematically evaluated now). Also, the field 'FLOPs' should be included.

@kennymckormick kennymckormick merged commit 570babe into open-mmlab:master Jul 1, 2021
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.

more pretrained checkpoints for ipCSN irCSN
2 participants