Push to main

[Rishit-dagli]

This PR pushes our existing work to main which adds the:

• Python Package

The Python package file can be present at ./tfwatcher, the two Firebase files in this directory help connect and write data to Firebase. The ./tfwatcher/callbacks subpackage creates new TensorFlow callback classes which can be used very easily as showed in docs.

This package could be run locally using:

# Use pip install -e .[dev] to install with dev dependencies
pip install -e .

• Documentation

The documentation is created with Sphinx and is to be hosted on https://readthedocs.org/ . Currently, the documentation isn't hosted but you could view the latest version locally by downloading the HTML or the PDF version (generated by GitHub Actions and uploaded as an artifact).

All the documentation code can be found at ./docs which could easily be developed locally as HTML by running:

cd docs
pip install sphinx
pip install -r requirements.txt
make html

The documentation can also be generated as a LaTeX PDF by running:

sudo apt-get update -y && sudo apt-get install -y latexmk texlive-latex-recommended texlive-latex-extra texlive-fonts-recommended imagemagick
make latexpdf

• A web app

The web app, created with React, Chakra-UI and Recharts is present at ./webapp. There are two main screens in the app(./screen), one for getting the key generated by the python package and another to show the logs for it. The Firebase instance to authenticate the user and get realtime data is in ./firebase, and the screens folder mentioned above.

To run locally:

cd webapp
yarn install
yarn run start

Here are some screenshots of the web-app in action:

---

cc: @shreyaparadkar @harshgeek4coder

[ayushmaang]

Code looks great! Love the documentation and the pdf you made.

[ayushmaang]

LGTM