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

feat: MLflow logging #91

Merged
merged 5 commits into from
Mar 13, 2024
Merged

feat: MLflow logging #91

merged 5 commits into from
Mar 13, 2024

Conversation

2018007956
Copy link
Contributor

@2018007956 2018007956 commented Mar 11, 2024

Overview

  • MLflow 사용하여 모델 학습 시 사용되는 파라미터 및 매트릭 값 로깅
  • python_env.yaml 대신 poetry 사용하여 의존성 관리

Change Log

  • File Tree
model
|
| # MLflow settings
├── MLProject              # project's meta info about execution & environment
├── pyproject.toml         # manage dependencies (instead of python_env.yaml)
├── poetry.lock     
│
| # Model training codes
├── base
│   └── base_trainer.py    # parameters & metrics logging  

To Reviewer

  • /model 의 부모 폴더에서 Run 및 서버 실행
    • Run : mlflow run model --experiment-name [experiment name] --env-manager=local
    • 서버 띄우기 : mlflow ui
  • 자동으로 세세한 로깅을 해주는 Autolog는 pytorch.nn.Module을 지원하지 않기 때문에,
    저장되어야 할 파라미터들을 수동으로 지정해 주는 것이 필요합니다.
    모델 실험 시 변경되는 부분(ex. 모델 아키텍처 등)이 잘 로깅 되는지 확인해주세요.

Issue Tags

@2018007956 2018007956 self-assigned this Mar 11, 2024
@2018007956 2018007956 merged commit 5f8fe28 into dev/model Mar 13, 2024
2018007956 added a commit that referenced this pull request Apr 21, 2024
* feat: stock data transformation & model baseline (#76)

* feat: get stock data & rasterize

#75

* feat: codes for train

#75

* feat: model to pytorch template (#87)

* remove: baseline_model

#86

* feat: data_generate_codes

#86

* feat: pytorch-template

#86

* fix: model call

#86

* fix: 모델 학습 코드 수정 (#89)

* remove: useless files

#88

* feat: ignore folder where model results are saved

#88

* remove: mnist codes

#88

* feat: poetry settings

#88

* remove: useless codes

#88

* fix: print statement

#88

* remove: tensorboard logging

#88

---------

Co-authored-by: xxxtenta <classaen7@naver.com>

* feat: MLflow logging (#91)

* feat: add poetry packages

#82

* feat: MLflow logging

#82

* feat: ignore model's by-product

#82

* fix: code bugs

#82

* feat: poetry add glances

#82

* docs: add folders to gitignore file

* fix: change folder name explicitly
model to model_training

* docs: README.md

---------

Co-authored-by: Chaea Kim <chaea11s0@gmail.com>
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.

1 participant