https://github.com/cbfinn/maml

Code for "Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks"
https://github.com/cbfinn/maml

Last synced: 6 days ago
JSON representation

Code for "Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks"

• Host: GitHub
• URL: https://github.com/cbfinn/maml
• Owner: cbfinn
• License: mit
• Created: 2017-06-18T01:36:06.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2020-01-19T22:05:37.000Z (almost 5 years ago)
• Last Synced: 2024-11-30T01:04:45.218Z (13 days ago)
• Language: Python
• Size: 559 KB
• Stars: 2,561
• Watchers: 47
• Forks: 606
• Open Issues: 44
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-few-shot-meta-learning - code - official? (TF)

README

        
# Model-Agnostic Meta-Learning

This repo contains code accompaning the paper, 	[Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks (Finn et al., ICML 2017)](https://arxiv.org/abs/1703.03400). It includes code for running the few-shot supervised learning domain experiments, including sinusoid regression, Omniglot classification, and MiniImagenet classification.

For the experiments in the RL domain, see [this codebase](https://github.com/cbfinn/maml_rl).

### Dependencies

This code requires the following:

* python 2.\* or python 3.\*

* TensorFlow v1.0+

### Data

For the Omniglot and MiniImagenet data, see the usage instructions in `data/omniglot_resized/resize_images.py` and `data/miniImagenet/proc_images.py` respectively.

### Usage

To run the code, see the usage instructions at the top of `main.py`.

### Contact

To ask questions or report issues, please open an issue on the [issues tracker](https://github.com/cbfinn/maml/issues).