This folder contains code to train XR-Transformer+PINA models and reproduce experiments in "PINA: Leveraging Side Information in eXtreme Multi-label Classification via Predicted Instance Neighborhood Aggregation".

• Clone the repository and enter examples/pina directory.
• First create a virtual environment and then install dependencies by running the following command:

pip install libpecos pandas gdown urllib3==1.26.6 

If you're unfamiliar with Python virtual environments, check out the user guide.

• Install pyxclib
• Verify pytorch and CUDA:

 python -c "import torch; print('torch={}, cuda={}'.format(torch.__version__, torch.cuda.is_available()))"

To train and evaluate PINA+XR-Transformer model, run

chmod a+x ./scripts/*
DATASET="LF-Amazon-131K"
bash scripts/run_pina.sh ${DATASET}

Recommended platform for training: AWS p3.16xlarge instance or equivalent.

[1] XC Repo: http://manikvarma.org/downloads/XC/XMLRepository.html