This repository implements a Decoder-Only Transformer from scratch using Python and PyTorch. The goal is to build a transformer model that can generate text based on an input prompt by predicting the ...

In Part 1 we already discussed about Transformer Encoder (almost 2 months ady haha >..<), now lets move to Part 2 (finally) where the model actually generate the output sequence. Why was the ...

Transformer architecture: • 1. The Core Engine (Self-Attention): This mechanism allows the model to analyze every word in a sentence simultaneously to determine how they relate to one another and ...

Research codebase accompanying the Parallel Decoder Transformer (PDT) paper. The repository contains the architecture scaffold, dataset pipeline, inference/orchestration stack, and training code for a ...

Captioning an image involves using a combination of vision and language models to describe the image in an expressive and concise sentence. Successful captioning task requires extracting as much ...