PyTorch Memory Deep Dive: view, reshape, transpose, permute and the Contiguity Puzzle
A practical deep dive into how PyTorch tensors use storage, stride, views and contiguity.
Incomplete
Parameter Efficient Fine Tuning Notes
Explaination of different Paramete Efficient Finetuning Techniques
Notes on PyTorch's Distributed Data Parallel (DDP)
A brief summary of PyTorch's implementation of Distributed Data Parallel DDP.
Zero Redunduncy Optimizer (ZeRO): Paper Summary
A brief note on ZeRO's workings.
Distributed training technologies for Transformers: Overview
A brief summary of different distributed training strategies used to train LLMs.
Named Entity Recognition (NER) as Machine Reading Comprehension (MRC)
In this article, I will give you a brief overview of Named Entity Recognition (NER), its importance in information extraction, few approaches to perform NER, and at the end will also show you how to implement NER as an MRC problem.
Train BERT for Question Answering Task
Let us dive into the BERT's architecture and details of formulating Question Answering NLP task for transformer models.
Abstractive Text Summarization with GPT2
In this article I will discuss an efficient abstractive text summarization approach using GPT-2 on PyTorch with the CNN/Daily Mail dataset.
How to Improve YOLOv3
YOLO has been a very popular and fast object detection algorithm, but unfortunately not the best-performing. In this article I will highlight simple training heuristics and small architectural changes that can make YOLOv3 perform better than models like Faster R-CNN and Mask R-CNN.