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] add log collector #1175

Merged
merged 8 commits into from
Jan 14, 2022
Merged

Conversation

RockeyCoss
Copy link
Contributor

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 a develop tool: Log Collector

Modification

Add a develop tool: Log Collector

Copy link
Collaborator

@MeowZheng MeowZheng left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

add copyright

RockeyCoss and others added 3 commits January 10, 2022 11:55
Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
@RockeyCoss RockeyCoss requested a review from MeowZheng January 10, 2022 04:16
Copy link
Collaborator

@MeowZheng MeowZheng left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@Junjun2016
Copy link
Collaborator

Hi @MengzhangLI
Please review this PR and give some comments.

Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>
@MengzhangLI
Copy link
Contributor

This enhancement could be ran correctly.

But the metric of all logs should keep the same, otherwise, the error may happens.

For example, if metric of log.json files have mDice and mIoU at the same time, but the input is mDice, then error may be raised during handling certain log.json file whose metric is mIoU.

image

However, if all .log.json files have same metric, it could run correctly.

image

lr_in_trans.json.md

exp_num method mDice best best index mDice last last index last iter num
1 deeplabv3_unet_s5-d16_ce-1.0-dice-3.0_128x128_40k_chase-db1 0.8949 10 0.8949 10 40000
2 deeplabv3_unet_s5-d16_ce-1.0-dice-3.0_128x128_40k_stare 0.9065 10 0.9065 10 40000
3 deeplabv3_unet_s5-d16_ce-1.0-dice-3.0_256x256_40k_hrf 0.8961 8 0.8956 10 40000
4 deeplabv3_unet_s5-d16_ce-1.0-dice-3.0_64x64_40k_drive 0.8892 7 0.8884 10 40000
5 fcn_unet_s5-d16_ce-1.0-dice-3.0_128x128_40k_chase-db1 0.8952 10 0.8952 10 40000
6 fcn_unet_s5-d16_ce-1.0-dice-3.0_128x128_40k_stare 0.9065 10 0.9065 10 40000
7 fcn_unet_s5-d16_ce-1.0-dice-3.0_256x256_40k_hrf 0.8964 10 0.8964 10 40000
8 fcn_unet_s5-d16_ce-1.0-dice-3.0_64x64_40k_drive 0.8897 4 0.8871 10 40000
9 pspnet_unet_s5-d16_ce-1.0-dice-3.0_128x128_40k_chase-db1 0.8945 10 0.8945 10 40000
10 pspnet_unet_s5-d16_ce-1.0-dice-3.0_128x128_40k_stare 0.9072 10 0.9072 10 40000
11 pspnet_unet_s5-d16_ce-1.0-dice-3.0_256x256_40k_hrf 0.8969 10 0.8969 10 40000
12 pspnet_unet_s5-d16_ce-1.0-dice-3.0_64x64_40k_drive 0.889 4 0.8876 10 40000

@MengzhangLI MengzhangLI added the Merging PR waited for merging label Jan 11, 2022
@RockeyCoss RockeyCoss requested a review from Junjun2016 January 11, 2022 16:12
@Junjun2016 Junjun2016 merged commit ba52d50 into open-mmlab:master Jan 14, 2022
@MengzhangLI MengzhangLI removed the Merging PR waited for merging label Jan 18, 2022
bowenroom pushed a commit to bowenroom/mmsegmentation that referenced this pull request Feb 25, 2022
* [Feature] add log collector

* Update .dev/log_collector/readme.md

Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>

* Update .dev/log_collector/example_config.py

Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>

* fix typo and so on

* modify readme

* fix some bugs and revise the readme.md

* more elegant

* Update .dev/log_collector/readme.md

Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>

Co-authored-by: Miao Zheng <76149310+MeowZheng@users.noreply.github.com>
Co-authored-by: Junjun2016 <hejunjun@sjtu.edu.cn>
aravind-h-v pushed a commit to aravind-h-v/mmsegmentation that referenced this pull request Mar 27, 2023
Flax tests: don't hardcode number of devices.

This makes it possible to test on CPU/GPU. However, expected slices are
only checked when there are 8 devices.
wjkim81 pushed a commit to wjkim81/mmsegmentation that referenced this pull request Dec 3, 2023
* clean inference code

* LoadImageFromFile supports given img
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.

4 participants