A Faster Pytorch Implementation of Multi-Head Self-Attention

News Articles Recommendation System. (Project for Practical Seminar in Machine Learning - PSI:ML 9)

gpt∀ - A modular, from-scratch implementation of the GPT architecture in PyTorch, covering attention, transformer blocks, and core LLM components.

Notebook used for the Kaggle Competition on Nasa-Cmapss dataset for the RUL predictions

My pytorch implemented solution to the Fall 2020 UC Berkley CS198 ViT homework; (P.S this is my first experience with ViTs let alone transformers please leave feedback!)