https://github.com/nasaharvest/timl

Task-Informed Meta-Learning
https://github.com/nasaharvest/timl

Last synced: 5 days ago
JSON representation

Task-Informed Meta-Learning

• Host: GitHub
• URL: https://github.com/nasaharvest/timl
• Owner: nasaharvest
• Created: 2022-01-31T12:52:59.000Z (almost 3 years ago)
• Default Branch: main
• Last Pushed: 2022-12-15T16:05:23.000Z (almost 2 years ago)
• Last Synced: 2024-08-04T03:11:25.429Z (4 months ago)
• Language: Python
• Homepage: https://arxiv.org/abs/2202.02124
• Size: 12.4 MB
• Stars: 16
• Watchers: 2
• Forks: 4
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Task-Informed Meta-Learning

This repository contains examples of Task-Informed Meta-Learning ([paper](https://arxiv.org/abs/2202.02124)).

We consider two tasks:

- [Crop Type Classification](crop_classification)

- [Yield Estimation](yield)

- [Grouped Omniglot](omniglot)

Each task acts as its own self-contained codebase - for more details on running the experiments, please check their respective READMEs.

### Getting started

For both tasks, [anaconda](https://www.anaconda.com/download/#macos) running python 3.6 is used as the package manager. To get set up with an environment, install Anaconda from the link above, and (from either of the directories) run

```bash

conda env create -f environment.yml

```

Once the environment is activated, the main script to train the models is then `deep_learning.py`, with the model configurations controlled by the `config.py` file.

The trained TIML models are available on [Zenodo](https://zenodo.org/record/6089828).