Skip to content

[MMSIG-88] Migrate face2d wingloss to 1.x #2297

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.

Sign up for GitHub

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

Jump to bottom
Merged
merged 1 commit into from
Apr 24, 2023
Merged

[MMSIG-88] Migrate face2d wingloss to 1.x #2297

merged 1 commit into from
Apr 24, 2023

Conversation

xin-li-67
Copy link
Contributor

@xin-li-67 xin-li-67 commented Apr 24, 2023 •
edited
Loading

Motivation

MMSIG-88: Migrate face2d wingloss to 1.x

Modification

configs/face_2d_keypoint/topdown_regression/README.md
configs/face_2d_keypoint/topdown_regression/wflw/td-reg_res50_wingloss_8xb64-210e_wflw-256x256.py
configs/face_2d_keypoint/topdown_regression/wflw/resnet_wingloss_wflw.md
configs/face_2d_keypoint/topdown_regression/wflw/resnet_wingloss_wflw.yml

BC-breaking (Optional)

Use cases (Optional)

Checklist

Before PR:

  • I have read and followed the workflow indicated in the CONTRIBUTING.md to create this PR.
  • Pre-commit or linting tools indicated in CONTRIBUTING.md are used to fix the potential lint issues.
  • Bug fixes are covered by unit tests, the case that causes the bug should be added in the unit tests.
  • New functionalities are covered by complete unit tests. If not, please add more unit tests to ensure correctness.
  • The documentation has been modified accordingly, including docstring or example tutorials.

After PR:

  • CLA has been signed and all committers have signed the CLA in this PR.

@xin-li-67
Copy link
Contributor Author

Test result using the new config file with 0.x checkpoint deeppose_res50_wflw_256x256_wingloss-f82a5e53_20210303.pth :

04/24 12:00:59 - mmengine - WARNING - The prefix is not set in metric class NME.
Loads checkpoint by local backend from path: work_dirs/td-reg_res50_wingloss_8xb64-210e_wflw-256x256/deeppose_res50_wflw_256x256_wingloss-f82a5e53_20210303.pth
04/24 12:00:59 - mmengine - INFO - Load checkpoint from work_dirs/td-reg_res50_wingloss_8xb64-210e_wflw-256x256/deeppose_res50_wflw_256x256_wingloss-f82a5e53_20210303.pth
/home/xinli/Projects/mmpose/mmpose/datasets/transforms/common_transforms.py:70: UserWarning: Use the existing "bbox_center" and "bbox_scale". The padding will still be applied.
  warnings.warn('Use the existing "bbox_center" and "bbox_scale"'
/home/xinli/Projects/mmpose/mmpose/datasets/transforms/common_transforms.py:70: UserWarning: Use the existing "bbox_center" and "bbox_scale". The padding will still be applied.
  warnings.warn('Use the existing "bbox_center" and "bbox_scale"'
04/24 12:01:04 - mmengine - INFO - Epoch(test) [50/79]    eta: 0:00:03  time: 0.103730  data_time: 0.044139  memory: 564  
04/24 12:01:07 - mmengine - INFO - Evaluating NME...
04/24 12:01:07 - mmengine - INFO - Epoch(test) [79/79]    NME: 0.046717  data_time: 0.045864  time: 0.088427

@xin-li-67
Copy link
Contributor Author

the logs of the first few training epochs:

2023/04/23 14:46:04 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io
2023/04/23 14:46:04 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future.
2023/04/23 14:46:04 - mmengine - INFO - Checkpoints will be saved to /home/xinli/Projects/mmpose/work_dirs/td-reg_res50_wingloss_8xb64-210e_wflw-256x256.
2023/04/23 14:46:16 - mmengine - INFO - Epoch(train)   [1][ 50/118]  lr: 6.193637e-06  eta: 1:38:58  time: 0.240141  data_time: 0.089458  memory: 7279  loss: 386.455813  loss_kpt: 386.455813  acc_pose: 0.005618
2023/04/23 14:46:27 - mmengine - INFO - Epoch(train)   [1][100/118]  lr: 1.244990e-05  eta: 1:34:58  time: 0.221637  data_time: 0.095511  memory: 7279  loss: 236.334516  loss_kpt: 236.334516  acc_pose: 0.034459
2023/04/23 14:46:31 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:46:42 - mmengine - INFO - Epoch(train)   [2][ 50/118]  lr: 2.095842e-05  eta: 1:33:51  time: 0.229608  data_time: 0.103010  memory: 7279  loss: 97.911779  loss_kpt: 97.911779  acc_pose: 0.128683
2023/04/23 14:46:54 - mmengine - INFO - Epoch(train)   [2][100/118]  lr: 2.721468e-05  eta: 1:33:08  time: 0.223200  data_time: 0.095928  memory: 7279  loss: 70.478111  loss_kpt: 70.478111  acc_pose: 0.272675
2023/04/23 14:46:57 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:47:09 - mmengine - INFO - Epoch(train)   [3][ 50/118]  lr: 3.572320e-05  eta: 1:32:15  time: 0.224575  data_time: 0.098417  memory: 7279  loss: 45.006643  loss_kpt: 45.006643  acc_pose: 0.394393
2023/04/23 14:47:20 - mmengine - INFO - Epoch(train)   [3][100/118]  lr: 4.197946e-05  eta: 1:31:49  time: 0.222053  data_time: 0.092891  memory: 7279  loss: 38.782641  loss_kpt: 38.782641  acc_pose: 0.528895
2023/04/23 14:47:24 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:47:35 - mmengine - INFO - Epoch(train)   [4][ 50/118]  lr: 5.048798e-05  eta: 1:31:15  time: 0.224032  data_time: 0.096709  memory: 7279  loss: 33.982139  loss_kpt: 33.982139  acc_pose: 0.526951
2023/04/23 14:47:46 - mmengine - INFO - Epoch(train)   [4][100/118]  lr: 5.674424e-05  eta: 1:30:53  time: 0.220662  data_time: 0.090002  memory: 7279  loss: 32.380127  loss_kpt: 32.380127  acc_pose: 0.473289
2023/04/23 14:47:49 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:48:01 - mmengine - INFO - Epoch(train)   [5][ 50/118]  lr: 6.250000e-05  eta: 1:30:33  time: 0.229566  data_time: 0.102327  memory: 7279  loss: 31.069123  loss_kpt: 31.069123  acc_pose: 0.738956
2023/04/23 14:48:12 - mmengine - INFO - Epoch(train)   [5][100/118]  lr: 6.250000e-05  eta: 1:30:29  time: 0.227160  data_time: 0.096115  memory: 7279  loss: 28.884894  loss_kpt: 28.884894  acc_pose: 0.690694
2023/04/23 14:48:17 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:48:28 - mmengine - INFO - Epoch(train)   [6][ 50/118]  lr: 6.250000e-05  eta: 1:30:48  time: 0.236433  data_time: 0.100837  memory: 7279  loss: 28.599137  loss_kpt: 28.599137  acc_pose: 0.544769
2023/04/23 14:48:39 - mmengine - INFO - Epoch(train)   [6][100/118]  lr: 6.250000e-05  eta: 1:30:24  time: 0.218485  data_time: 0.091582  memory: 7279  loss: 26.704508  loss_kpt: 26.704508  acc_pose: 0.723798
2023/04/23 14:48:43 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:48:55 - mmengine - INFO - Epoch(train)   [7][ 50/118]  lr: 6.250000e-05  eta: 1:30:09  time: 0.226730  data_time: 0.099920  memory: 7279  loss: 26.470955  loss_kpt: 26.470955  acc_pose: 0.732419
2023/04/23 14:49:06 - mmengine - INFO - Epoch(train)   [7][100/118]  lr: 6.250000e-05  eta: 1:29:46  time: 0.217243  data_time: 0.091017  memory: 7279  loss: 25.537712  loss_kpt: 25.537712  acc_pose: 0.618997
2023/04/23 14:49:09 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:49:21 - mmengine - INFO - Epoch(train)   [8][ 50/118]  lr: 6.250000e-05  eta: 1:29:24  time: 0.225552  data_time: 0.099602  memory: 7279  loss: 24.596247  loss_kpt: 24.596247  acc_pose: 0.724618
2023/04/23 14:49:32 - mmengine - INFO - Epoch(train)   [8][100/118]  lr: 6.250000e-05  eta: 1:29:06  time: 0.218679  data_time: 0.091527  memory: 7279  loss: 24.544875  loss_kpt: 24.544875  acc_pose: 0.628463
2023/04/23 14:49:35 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:49:46 - mmengine - INFO - Epoch(train)   [9][ 50/118]  lr: 6.250000e-05  eta: 1:28:43  time: 0.223306  data_time: 0.096661  memory: 7279  loss: 26.514483  loss_kpt: 26.514483  acc_pose: 0.649837
2023/04/23 14:49:48 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:49:57 - mmengine - INFO - Epoch(train)   [9][100/118]  lr: 6.250000e-05  eta: 1:28:28  time: 0.220764  data_time: 0.094428  memory: 7279  loss: 25.391091  loss_kpt: 25.391091  acc_pose: 0.607339
2023/04/23 14:50:01 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:50:13 - mmengine - INFO - Epoch(train)  [10][ 50/118]  lr: 6.250000e-05  eta: 1:28:10  time: 0.226316  data_time: 0.099465  memory: 7279  loss: 24.439167  loss_kpt: 24.439167  acc_pose: 0.737695
2023/04/23 14:50:23 - mmengine - INFO - Epoch(train)  [10][100/118]  lr: 6.250000e-05  eta: 1:27:49  time: 0.213511  data_time: 0.085391  memory: 7279  loss: 24.544752  loss_kpt: 24.544752  acc_pose: 0.777562
2023/04/23 14:50:27 - mmengine - INFO - Exp name: td-reg_res50_wingloss_8xb64-210e_wflw-256x256_20230423_144601
2023/04/23 14:50:27 - mmengine - INFO - Saving checkpoint at 10 epochs

@codecov
Copy link

codecov bot commented Apr 24, 2023

Codecov Report

Patch and project coverage have no change.

Comparison is base (98d0dfa) 82.07% compared to head (999ad5f) 82.07%.

❗ Current head 999ad5f differs from pull request most recent head 654c28b. Consider uploading reports for the commit 654c28b to get more accurate results

Additional details and impacted files 
@@           Coverage Diff            @@
##           dev-1.x    #2297   +/-   ##
========================================
  Coverage    82.07%   82.07%           
========================================
  Files          232      232           
  Lines        13643    13643           
  Branches      2319     2319           
========================================
+ Hits         11197    11198    +1     
+ Misses        1913     1912    -1     
  Partials       533      533
Flag Coverage Δ
unittests 82.07% <ø> (+<0.01%) ⬆️

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

see 1 file with indirect coverage changes

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 in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

@Tau-J Tau-J merged commit d79a2ed into open-mmlab:dev-1.x Apr 24, 2023
Tau-J pushed a commit to Tau-J/mmpose that referenced this pull request Apr 25, 2023
@xin-li-67 @Tau-J
migrate init (open-mmlab#2297)
29fe385
shuheilocale pushed a commit to shuheilocale/mmpose that referenced this pull request May 6, 2023
@xin-li-67
migrate init (open-mmlab#2297)
28c1a14
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Tau-J Tau-J Tau-J approved these changes

Assignees

@Tau-J Tau-J

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@xin-li-67 @Tau-J