ML-DL

Transformer architecture variation: RMSNorm 05-11

One for all: the torch.einsum API 04-14

Weight Tying in Language Models: A Technique to Parameter efficiency 03-11

What is Multi-Head Attention (MHA) 03-04

An Explanation of Self-Attention mechanism in Transformer 03-02

The Flow of GraphRAG 02-12

How KNN Algorithm Works 12-15

LLM inference optimization - KV Cache 10-12

LoRA fine-tuning 09-14

A trick to calculating partial derivatives in machine learning 07-26

Demystifying Pytorch's Strides Format 07-14

How to understand the backpropagation algorithm 04-04

Linear Regression Model Guide - theory part 03-15