Skip to content

Latest commit

 

History

History
51 lines (34 loc) · 829 Bytes

README.md

File metadata and controls

51 lines (34 loc) · 829 Bytes

mlops_demo

Manage machine learning experiments using MLflow and Hydra.

Install

$ git clone https://github.com/futabato/mlops_demo
$ cd mlops_demo/

pip install

$ pip install -r requirements.txt

docker

$ docker build -t mlops_demo .
$ docker run --rm -it -v "${PWD}:/home" --name mlops_demo -p 5000:5000 -p 8888:8888 mlops_demo /bin/bash

Hands-on

  • mlflow/
    • mlflowを説明するためのコード。
  • hydra/
    • hydraを説明するためのコード。
  • Integration/
    • mlflowとhydraを組み合わせた実験管理を説明するためのコード。

Launching the Tracking UI

$ mlflow ui

or

$ mlflow ui -h `hostname`

access http://localhost:5000

Slide

https://speakerdeck.com/futabato/mlflowtohydrawoli-yong-sitashi-yan-guan-li