flukeskywalker/nanoDD

Simple Scalable Discrete Diffusion for text in PyTorch

23
/ 100
Experimental

This project helps machine learning researchers and practitioners understand and implement Discrete Diffusion models for text generation. Unlike traditional language models that generate text word-by-word, this approach generates text in parallel, starting from a noisy sequence and iteratively refining it. It takes raw text data as input and produces newly generated text sequences, serving as a foundational tool for those exploring advanced text generation techniques.

No commits in the last 6 months.

Use this if you are a machine learning researcher or engineer interested in experimenting with, learning about, or building upon discrete diffusion models for text generation.

Not ideal if you need a production-ready, off-the-shelf text generation tool for immediate application without deep dives into model architecture or training.

text-generation natural-language-processing machine-learning-research diffusion-models deep-learning
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 8 / 25
Community 8 / 25

How are scores calculated?

Stars

37

Forks

3

Language

Python

License

Last pushed

Sep 27, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/flukeskywalker/nanoDD"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.