https://github.com/juftin/fastapp

HTTP Servers Made Easy - Optimized for ML
https://github.com/juftin/fastapp

asgi fastapi machine-learning

Last synced: about 2 months ago
JSON representation

HTTP Servers Made Easy - Optimized for ML

• Host: GitHub
• URL: https://github.com/juftin/fastapp
• Owner: juftin
• License: mit
• Created: 2022-02-04T15:53:24.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2023-02-01T14:01:22.000Z (almost 2 years ago)
• Last Synced: 2024-11-15T04:59:01.939Z (about 2 months ago)
• Topics: asgi, fastapi, machine-learning
• Language: Python
• Homepage: https://juftin.com/fastapp/
• Size: 3.38 MB
• Stars: 4
• Watchers: 2
• Forks: 2
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# FastApp

HTTP Apps Made Easier with FastApp

## Installation

```shell

pip install fastapp

```

## Using Out the Example Server

```shell

pip install fastapp[example]

```

```shell

fastapp serve-debug fastapp.app.example:app

```

## Using FastApp to build an app

Create a Python File with Endpoints, we'll call this `main.py`:

```python

from datetime import datetime

from fastapp.app import app

@app.get("/hello")

def custom_endpoint() -> dict:

    """"

    This is a Custom API Endpoint

    """

    return dict(timestamp=datetime.now(),

                hello="world")

```

Then, using the `FastApp` CLI we can serve this App:

```shell

fastapp serve-debug main:app

```

...or via docker:

```shell

docker run --rm -it \

    --publish 8080:8080 \

    --volume ${PWD}/main.py:/root/fastapp/main.py \

    juftin/fastapp:latest \

    serve-debug main:app

```

Test out our new endpoint:

```shell

curl \

  --request GET \

  --header "Content-Type: application/json" \

  http://localhost:8080/hello

```

Alternatively, if we want to serve this app using Gunicorn, Nginx, and the UvicornWorker we can use

the `serve` command:

```shell

fastapp serve main:app

```

I prefer doing this within a docker container so you don't have to run Nginx on the host machine:

```shell

docker run --rm -it \

    --publish 8080:8080 \

    --volume ${PWD}/main.py:/root/fastapp/main.py \

    juftin/fastapp:latest \

    serve main:app

```