jakesnell/prototypical-networks
PyTorch implementation of prototypical networks for few-shot learning, a metric-learning approach to classification from few examples.
Not currently ranked — collecting fresh signals.
star history
This repository provides code for the prototypical networks paper (NeurIPS 2017), implementing an embedding-based few-shot classification method. The approach learns a metric space where class prototypes are computed from support examples, enabling classification from few labeled samples. It includes training and evaluation scripts with support for GPU acceleration and train/validation workflows.
Frequently asked
- What is jakesnell/prototypical-networks?
- PyTorch implementation of prototypical networks for few-shot learning, a metric-learning approach to classification from few examples.
- Is prototypical-networks open source?
- Yes — jakesnell/prototypical-networks is open source, released under the MIT license.
- What language is prototypical-networks written in?
- jakesnell/prototypical-networks is primarily written in Python.
- How popular is prototypical-networks?
- jakesnell/prototypical-networks has 1.2k stars on GitHub.
- Where can I find prototypical-networks?
- jakesnell/prototypical-networks is on GitHub at https://github.com/jakesnell/prototypical-networks.