/fastapi-ml-template

Primary LanguagePythonMIT LicenseMIT

FastAPI ML Template

I wrote a blog post about this repository in Japanese.

機械学習の推論WebAPIの実装をテンプレート化して使い回せるようした

Run Web API

Local

$ sh run.sh
$ poetry run uvicorn app.main:app --reload --host 0.0.0.0 --port 9000

Docker

$ docker build -f Dockerfile -t fastapi-ml .
$ docker run -p 9000:9000 --rm --name fastapi-ml -t -i fastapi-ml

Docker Compose

$ docker compose up --build

Request Commands

$ curl --request POST --url http://127.0.0.1:9000/api/v1/predict --header 'Content-Type: application/json' --data '{"input_text": "test"}'
$ http POST http://127.0.0.1:9000/api/v1/predict input_text=テスト

Development

Run Tests and Linter

$ poetry run tox

Reference