This is a PyTorch implementation of our paper "Multi-Instance Partial-Label Learning: Towards Exploiting Dual Inexact Supervision. Science China Information Sciences, in press."
Authors: Wei Tang, Weijia Zhang, and Min-Ling Zhang
@article{tang2023mipl,
title={Multi-Instance Partial-Label Learning: Towards Exploiting Dual Inexact Supervision},
author={Wei Tang and Weijia Zhang and Min-Ling Zhang},
journal={Science China Information Sciences},
year={2023}
}
gpytorch==1.8.0
numpy==1.21.5
scipy==1.7.3
torch==1.12.0
To install the requirement packages, please run the following command:
pip install -r requirements.txt
The datasets used in this paper can be found on this link.
To reproduce the results of MNIST_MIPL dataset in the paper, please run the following command:
bash demo.sh
This package is only free for academic usage. Have fun!