jakesnell/prototypical-networks

Code for the NeurIPS 2017 Paper "Prototypical Networks for Few-shot Learning"

51
/ 100
Established

This code helps machine learning researchers and practitioners explore "few-shot learning" by implementing Prototypical Networks. It takes in small datasets (like Omniglot) where each class has only a few examples, and outputs a trained model capable of classifying new, unseen examples from those classes. This is for researchers working on novel machine learning approaches where data is scarce for many categories.

1,220 stars. No commits in the last 6 months.

Use this if you are a machine learning researcher or student investigating few-shot learning techniques and want a reference implementation of Prototypical Networks.

Not ideal if you need an out-of-the-box solution for a production application or require extensive hyperparameter tuning and model management features.

few-shot learning meta-learning machine learning research model training pattern recognition
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

1,220

Forks

271

Language

Python

License

MIT

Last pushed

Jan 28, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/jakesnell/prototypical-networks"

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