Aman Kumar Raj's Projects
This is a unity package, which creates 3D text in unity.
Config files for my GitHub profile.
This repository contains the code for our 2nd Place solution to the SDS BIT Mesra - ML Contest on Liver Disease Prediction hosted on DPhi.
In this project, We use transfer learning to create a Cat or not Cat binary classifier.
Measure execution time for each line of a code.
This is a 3D game made with unity. In this person has to move on bridge without colliding with different color obstacle along with collecting coins.
Language Model that creates catchy, exciting, innovative, captivating, creative and engaging captions instead of just a description of the picture.
Through this project we will try to understand CutMix by implementing it on a simple problem of cat-vs-dog classification.
Implementation of EfficientNet model. Keras and TensorFlow Keras.
Find optimal path for electric vehicle using spatial information.
This is a first aid app which provides accurate first aid regarding given emergency condition.
Framework3 is a super-simple and robust ML Pipeline for tabular and image competition. The purpose of this is to make the process not too abstract, so that the user can have full control over it.
This repository helps find most complex repository of a user not just based on file structure but rather code complexity presence of nested loops , coupling and cohesions between classes and modules etc.
This is a unity game based on halloween night theme.
In this project we use Reinforcement Learning to extract features from an image.
Empowering Effortless Data Linking
This repository contains Machine Learning algorithms implementation from scratch.
It is a classical Reinforcement Learning problem.
Through this project we will try to understand working of Steepest-Descent and Gradient-Descent method and the differences between them.
This project is part of Product Recommender Hackathon of UNIV.ai.
SciPy library main repository
In this project I built a search tool (a small retrieval system) using Boolean retrieval method for cranfield dataset. and apply various queries with conjunction, disjunction and negation of terms. Here we also have to see the effectiveness of our implementation using MAP and Precision Recall graph for 1.without stemming 2.with stemming 3. zone indexes for title and body text fields and come up with optimal weights.
Tez is a super-simple and lightweight Trainer for PyTorch. It also comes with many utils that you can use to tackle over 90% of deep learning projects in PyTorch.
Tabular Playground Series Kaggle Competitions
Tabular Playground Series-Oct 2021
Uninformed search is a class of general-purpose search algorithms that operates in brute force-way. Here we have solved two problems Missionary and Cannibal Problem, Rabbit Leap Problem.
In this project, We will learn to work with different file formats like csv,json,xml,ods and many more.