sahiljain314 Goto Github PK

autozoom

Automatically get to Zoom class

barebones-machine-learning-course

civitas

Discord bot that uses deep learning to detect and blocks harassing messages. Second place LancerHacks

classifeye

Staying awake to finish a project can be very hard. Not with classifeye! We use CNNs to detect your eyes through your webcam and classify them into open/closed. If your eyes stay closed for too long, we play a loud alarm to wake you up.

curio

devilsadvocate

We often get trapped in our respective echo chambers when it comes to the news media. DevilsAdvocate aims to help people break out of this. We made a chrome extension that implements a Deep learning based classification system to detect bias in the news website of one's choosing. After reporting this bias, DevilsAdvocate will find articles that have the opposite bias to present to you. With this, you can get a balanced news stream and become better informed. Second place at PalyHacks

flax

Flax is a neural network library for JAX that is designed for flexibility.

gamedevelopmentlevelgame

CS498GD LevelGame MP

gamedevelopmentmp1

Machine Problem 1 for CS498GD

glass-eye-android

2nd Place LaunchHacks 2019

gradinator

An automatic differentiation and neural network library created from core principles and math in numpy for education.

malariacells

This is a Convolutional Neural Network image classification system with a flask web-app and android front end for detecting malaria in pictures of cells. 1st place MatadorHacks

pure_patents

rectifiedflowparallel

Rectified Flows improve Parallel Sampling

sahiljain314.github.io

suspect-sketch

People often have trouble recounting the details of faces that they have seen and this is unfortunate at places like crime scenes. To help this, we created a neural network that can guide generations from a Generative Adversarial Network(StyleGAN2) so that we get a realistic depiction of a potential suspect's face. First Place, DVHacks II

t5x

webdataset

A high-performance Python-based I/O system for large (and small) deep learning problems, with strong support for PyTorch.