Harshit Varma's Projects
Repository containing code of Team AttesTable at Task 9 SemEval 2021
A MAP-MRF Framework for Image Denoising
A Bayesian framework for Multi-Frame Image Super-Resolution. Based on "Bayesian Image Super-Resolution" (ME Tipping and CM Bishop, NeurIPS 2003)
A few cellular automatons implemented in C++ with an interactive GUI made using FLTK
The classic snake game implemented in C++ using the simplecpp package
[ICML 2023] Official code for our paper: 'Conditional Tree Matching for Inference-Time Adaptation of Tree Prediction Models'
Contains all related code for the Deep Learning Specialization on Coursera, offered by deeplearning.ai
Estimating various properties of globular clusters from astrophotometric data
Image Segmentation using Fuzzy C-Means Clustering with Bias Field Correction
POS tagging using a Hidden Markov Model (HMM) with Viterbi Decoding
Image segmentation using the EM algorithm that relies on a GMM for intensities and a MRF model on the labels. Based on "Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm" (Zhang, Y et al.)
My GitHub profile README
Contains all related code for the Machine Learning course on Coursera, offered by Stanford University.
Some Machine Learning algorithms implemented by me, mostly from scratch
An implementation of "Exact Maximum A Posteriori Estimation for Binary Images" (D. Greig, B. Porteous and A. Seheult)
Named Entity Identification (NEI) using SVM
A PyTorch implementation of Legendre Memory Units (LMUs) and its FFT variant
Region quadtrees for operations on binary images, implemented in C++
Character-Level LSTM used for generating Shakespearean text.
A python turtle that sketches (almost) any image given to it, while mimicking the "human way" of sketching
Statistical analysis of 2D shapes based on "Generalized Procrustes Analysis" (Gower, J.C.) and "Active Shape Models - Their Training and Application" (Cootes et al.)
End-to-End Conditional Poetry Generation
BiLSTM with Multi-Headed Self Attention for sentiment classification of Twitter data, implemented in Keras and PyTorch.
Word Sense Disambiguation using the Lesk algorithm with Word2Vec embeddings