dragen1860/MAML-Pytorch

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

2483 stars · 437 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch implementation of Model-Agnostic Meta-Learning (MAML) for few-shot supervised learning, supporting MiniImagenet and Omniglot benchmarks. It serves as a starting point for meta-learning research rather than a production library.

Use cases

• reproduce MAML few-shot learning experiments in PyTorch
• run 5-way 1-shot classification on MiniImagenet
• train meta-learner on Omniglot dataset
• learn how meta-learning inner/outer loops are implemented
• baseline code for few-shot learning research

When to choose

• you want a readable PyTorch reference implementation of MAML
• you need a starting point for few-shot meta-learning research
• you want to benchmark against MiniImagenet or Omniglot

When to avoid

• you need a maintained production library with active support
• you want first-order methods like Reptile (use the author's Reptile repo)
• you need TensorFlow instead of PyTorch

Facets

library · maturity maintenance

machine-learning deep-learning machine-learning deep-learning artificial-intelligence python meta-learning few-shot-learning pytorch maml research-code miniimagenet omniglot gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dragen1860/MAML-Pytorch")