First of all, congrats to Jeremy and Sylvain for a huge successful fastai version 2 launch.
To help fastai community with new version, I have also updated deployment ready templates for creating responsive web app for latest Fastai2 based Vision models. This started pack can be deployed to any docker based web service hosting services, some of them are:
https://azure.microsoft.com/en-us/services/app-service/containers/
https://aws.amazon.com/ecs/
https://cloud.google.com/cloud-build/docs/
https://www.digitalocean.com/products/one-click-apps/docker/
https://render.com
https://zeit.co/now
https://getcarina.com/
https://sloppy.io/en/
https://giantswarm.io
I have also updated statics file with few additional CSS & JS quirks to handle big camera file uploads. So, You can continue to test your fastai2 vision models from real world images from mobile devices.
I have previously wrote a detailed guide to build & deploy these starter pack as a web app on 5 Cloud services, including AWS Beanstalk, Google App Engine, Plus Azure Website, Render & now.sh.
Here is link =>
https://towardsdatascience.com/building-web-app-for-computer-vision-model-deploying-to-production-in-10-minutes-a-detailed-ec6ac52ec7e4
I hope to keep updating this repo and guide and create more starter packs for Text or Collab Filtering.
You can download starter pack here:
git clone https://github.com/psmathur/fastai2-vision-app
Please do let me know your questions & feedback.
Thanks,
Pankaj