bobf / audio-app-mf-ct

Audio Classification Web App using Freesound 2019 dataset

audio-app-mf-ct

Multi-Label Audio Classification Web App using kaggle Freesound 2019 curated dataset

Deplyed on https://render.com/

The guide for production deployment to Render is at https://course.fast.ai/deployment_render.html.

See Render's fast.ai forum thread for questions and support.

Heroku additions

Taken from: https://github.com/sachinchaturvedi93/PokemonClassifier with additions

1. The starter code taken from for deploying fast.ai models on Render. Plus additions for fastai2 and fastai2-audio in this repo.
2. Add the Procfile to the repository and put web: python app/server.py serve in it.
3. Add a runtime.txt file to specify python version.
4. In the server.py file we need to add :

import os 
import requests
Port = int(os.environ.get('PORT', 50000))

and replace uvicorn.run(.....) by

uvicorn.run(app=app, host='0.0.0.0', port=Port, log_level="info")

Edit : After upgrading uvicorn version, the app crashed with the following error - Publish to Heroku is broken: "WARNING: You must pass the application as an import string to enable 'reload' or 'workers"

Solution is changing WEB_CONCURRENCY = 1 in app's setting on Heroku dashboard.