https://github.com/ikostrikov/pytorch-meta-optimizer

A PyTorch implementation of Learning to learn by gradient descent by gradient descent
https://github.com/ikostrikov/pytorch-meta-optimizer

deep-learning gradient-descent pytorch

Last synced: about 1 month ago
JSON representation

A PyTorch implementation of Learning to learn by gradient descent by gradient descent

• Host: GitHub
• URL: https://github.com/ikostrikov/pytorch-meta-optimizer
• Owner: ikostrikov
• License: mit
• Created: 2017-01-30T05:26:28.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2018-08-27T09:04:29.000Z (over 5 years ago)
• Last Synced: 2024-01-25T20:08:54.929Z (4 months ago)
• Topics: deep-learning, gradient-descent, pytorch
• Language: Python
• Size: 17.6 KB
• Stars: 306
• Watchers: 16
• Forks: 56
• Open Issues: 6
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Lists

• Awesome-pytorch-list - Learning to learn by gradient descent by gradient descent
• Awesome-pytorch-list-CNVersion - Learning to learn by gradient descent by gradient descent

README

        
# Intro

PyTorch implementation of [Learning to learn by gradient descent by gradient descent](https://arxiv.org/abs/1606.04474).

## Run

```bash

python main.py

```

### TODO

- [x] Initial implementation

- [x] Toy data

- [x] LSTM updates

- [ ] Refactor, find a better way to organize the modules

- [ ] Compare with standard optimizers

- [x] Real data

- [ ] More difficult models