Tejas Morkar

Sketch to Color

Sketch to Color Image generation is an image-to-image translation model using Conditional Generative Adversarial Networks as described in the original paper by Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, Alexei A. Efros 2016, Image-to-Image Translation with Conditional Adversarial Networks.

The model was trained on the Anime Sketch-Colorization Pair Dataset available on Kaggle which contains 14.2k pairs of Sketch-Color Anime Images. The training of the model was done for 150 epochs which took approximately 23 hours on a single GeForce GTX 1060 6GB Graphic Card and 16 GB RAM.

ToxicityZero Discord Bot

Bot for helping reduce toxicity level on discord servers to zero!

This is a discord bot that I built while giving a Microsoft Learn Student Ambassadors presentation on how to build Discord bots using discord.js and tensorflow.js libraries. It was then deployed on Azure App Services using the continuous WebJobs for guaranteed uptime of the bot.

You can find the entire session recording and presentation slide deck on this link.

Crwn Clothing React App

A demo shopping web app built using React

You can Sign Up / Sign In using email address and password or google account, explore Hats, Jackets, and Sneakers for Women and Men, add items to cart, add/ remove items from checkout page, and pay using Stripe's test Visa card.

Streamlit App for Housing Price Prediction Tutorial

This is a Streamlit Web App that hosts a Housing Price prediction model that can be tweaked on the web itself. You can try to change number of neurons, learning rate, and epochs. It is an exploratory tutorial into the California Housing Price dataset and the building of the prediction model using TensorFlow in python.

If you are like me or any other Machine Learning enthusiast, the worst nightmare for you, obviously other than the Cuda errors, must be deploying the model elsewhere to present your work to your friends and networks on the internet. There are many ways to go around and achieve this, I think Streamlit is one of the go-to solution for it.

Sudoku Solver

This is a project which takes an input of the sudoku puzzles as an 81 characters long string containing the digits from 0 to 9, where 0 represents an unfilled cell of the Sudoku Board. The objective of the algorithm is to replace all the zeros with valid digits such that the entire Sudoku Puzzle is solved. For doing this in a conventional way, the processor will require to try all 2*1077 possible combinations in a worst case scenario which is a huge number. So, to solve this problem in a comparatively less time, the Backtracking Algorithm is one of the best solutions to it.

Sudoku is a logic-based, combinatorial number-placement puzzle. The objective is to fill a 9×9 grid with digits so that each column, each row, and each of the nine 3×3 subgrids that compose the grid contain all of the digits from 1 to 9. (Source - Wikipedia)