RahulSChand

gpu_poor

Calculate token/s & GPU memory requirement for any LLM. Supports llama.cpp/ggml/bnb/QLoRA quantization

llama2.c-for-dummies

Step by step explanation/tutorial of llama2.c

Multi-Granularity-Hierarchical-Attention-Fusion-Networks-for-Question-Answering---TensorFlow

TensorFlow implementation of the "Multi-Granularity Hierarchical Attention Fusion Networks for Reading Comprehension and Question Answering" paper by Alibaba.

Weighted-low-rank-factorization-Pytorch

PyTorch implementation of Language model compression with weighted low-rank factorization

batched-pytorch-ksvd

Batched version of KSVD written in PyTorch that runs on GPU

gpt2_squad

GPT2 training on squad dataset

llama2.c-omp-cuda

llama2.c with GPU offloading using OMP

Optical-Flow-Shape-Dataset

Code to create shape data set for optical flow tasks

rahulschand.github.io

Just a plain, simple and elegant one-page theme for research/academia.

rahulschand.github.io-old

Personal website made using NextJS & tailwindcss

ALL-LLM-Quant

Implementation & Visualisation of all (widely used) LLM quantisation methods

Compiler-Construction

EIC-Submission

EIC-Assignment

Expression-Parser-for-Prolog

IIRS-Vehicle-Detection

implicitCoT

llm-reasoners-eurus

A library for advanced large language model reasoning

MoE-Training-Vanilla

Vanilla Implementation of MoE training - Expert/Model/Data Parallelism

monosemanticity-quantization

Affect of Quantization on Monosemanticity of 1 layered Networks

PWC-Net

PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume, CVPR 2018 (Oral)