Using the Web App
=================

TF Watcher Webapp is built with React, Chakra-UI and Firebase SDK. The main aim of this 
application is to monitor ML training or testing process. It is connected with Firebase,
to get real time updates and show graphs for the same.

Try out the web app `here <https://www.tfwatcher.tech/>`_

.. only:: not latex

  .. image:: _static/webapp-screenshot.png
    :alt: Web App in Action

Features
--------

- Authentication
- Connecting with Firebase Realtime Database
- Read data and format it to display charts
- Deployed to Azure Static Webapp
- Tests (WIP)

Using the web app
-----------------

Using this web app is indeed quite simple, you would need to be logged in with Google 
and enter the ID generated by the Python package.

.. only:: not latex

  .. image:: _static/add-key.png
    :alt: Adding keys in Web App

You can also share the live page for your logs or the previous saved version with 
others.

Development setup
-----------------

You will require Yarn or npm to be able to run this application locally. Download npm 
from `here <https://docs.npmjs.com/downloading-and-installing-node-js-and-npm>`__ and 
Yarn from `here <https://classic.yarnpkg.com/en/docs/install#windows-stable)>`__ .

To setup ``tfwatcher webapp`` and get all its dependencies, run the following:

.. code-block:: bash

    git clone https://github.com/Rishit-dagli/TF-Watcher.git
    # or clone your own fork

    cd webapp
    yarn install

To run the application locally:

.. code-block:: bash

    yarn start

To create a build version:

.. code-block:: bash

    yarn run build