Peter (CTO) likes images but loves data augmentation more. He wants a way to see how typical image augmentation in our machine learning affects the images we input to the model. He codes a simple boiler-plate application for you and asks you to finish it off!
To complete the flask app you will need to do the following:
- Download (Do not fork) the repo, upload to a new repo in your name/account.
- Write a dockerfile for amd64 for the app.
- Use github actions to build the dockerfile.
- Add a image augmentation function image_augmentations.py
- Add a test for your function
- Improve the repo/python code/deployment in some imaginative way to show-off... Highlight this in the readme.
- Send the team the link to your code.
Put the instructions here on how to set up/run the docker file. It should port-forward the application to run on http://127.0.0.1:8000/.
To send an image and get the app to work open a link as follows: http://0.0.0.0:8000/?url=<image_url> for example: http://0.0.0.0:8000/?url=https://imgs.xkcd.com/comics/bad_code.png
Please find instructions in the sub-projects: mymodel and flask_augmentations.
