dpressel/mint
MinT: Minimal Transformer Library and Tutorials
This is a hands-on toolkit for machine learning engineers and researchers to build core Transformer models from the ground up. It provides clear tutorials and a minimal Python library to understand how models like BERT, GPT, and T5 work internally. You can start with raw text data, train these models, and then use them for tasks like text completion or classification.
261 stars. No commits in the last 6 months.
Use this if you are a machine learning engineer or researcher who wants to deeply understand and implement Transformer architectures for natural language processing from foundational principles.
Not ideal if you are looking for a high-level API to quickly apply pre-trained Transformer models without needing to build them yourself.
Stars
261
Forks
15
Language
Python
License
—
Category
Last pushed
Jul 26, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/dpressel/mint"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lucidrains/x-transformers
A concise but complete full-attention transformer with a set of promising experimental features...
kanishkamisra/minicons
Utility for behavioral and representational analyses of Language Models
lucidrains/simple-hierarchical-transformer
Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT
lucidrains/dreamer4
Implementation of Danijar's latest iteration for his Dreamer line of work
Nicolepcx/Transformers-in-Action
This is the corresponding code for the book Transformers in Action