MARMOT is a Python library for dealing with multimodal data consisting of both images and text. It does not require multimodal pretraining and it can deal with observations missing modalities. This repo is currently under construction and is in a very rough state; please feel free to reach out if you have any questions.
To use the functions in this package, install the package first. To do this, navigate to the directory you would like to install the files in. Then,
git clone https://github.com/patrickywu/MARMOT
cd MARMOT
pip install -r requirements.txt
python setup.py develop
- Patrick Y. Wu (University of Michigan)
- Walter R. Mebane, Jr. (University of Michigan)