Themeda

A model to forecast changes to ecosystems in Australia.

For more information, see the preprint here: https://ssrn.com/abstract=4681094

Installation

Install using pip:

pip install git+https://github.com/themeda-devs/themeda.git

Usage

See the options for training a model with the command:

themeda train --help

To use the ConvLSTM model, use the themeda-convlstm command:

themeda-convlstm train --help

See the options for making inferences with the command:

themeda infer --help

Credits

• Robert Turnbull
• Damien Mannion
• Jessie Wells
• Kabir Manandhar Shrestha
• Attila Balogh
• Rebecca Runting

If you use Themeda, please cite the preprint:

Turnbull, Robert and Mannion, Damien and Wells, Jessie and Manandhar Shrestha, Kabir and Balogh, Attila and Runting, Rebecca, Themeda: Predicting land cover change using deep learning (January 1, 2024). Available at SSRN: https://ssrn.com/abstract=4681094

@article {themeda2024,
    author = {Turnbull, Robert and Mannion, Damien and Wells, Jessie and Manandhar Shrestha, Kabir and Balogh, Attila and Runting, Rebecca},
    title = {{Themeda: Predicting land cover change using deep learning}},
    year = {2024},
    doi = {10.2139/ssrn.4681094},
    journal = {SSRN}
}