TPAMI 2023
- Python 3.10
- PyTorch 1.12.1
- faiss
conda install -c pytorch faiss-gpu
Modify the ./Utils/PathPresettingOperator.get_dataset_path, then train the model(s):
# NoisyMNIST
python main.py --dataset NoisyMNIST30000 --seed 9116 --aligned_prop 1 --complete_prop 1
# MNISTUSPS
python main.py --dataset MNISTUSPS --seed 9116 --aligned_prop 1 --complete_prop 1
# Caltech
python main.py --dataset 2view-caltech101-8677sample --seed 9116 --aligned_prop 1 --complete_prop 1
# CUB
python main.py --dataset cub_googlenet_doc2vec_c10 --seed 9116 --aligned_prop 1 --complete_prop 1
# YouTubeFaces
python main.py --dataset YouTubeFaces --seed 9116 --aligned_prop 1 --complete_prop 1
The pre-trained models are available here.
Download the models, then:
python main.py --dataset dataset --seed seed --resume PathToYourModel
If SMILE is useful for your research, please cite the following paper:
@article{zeng2023semantic,
title={Semantic Invariant Multi-view Clustering with Fully Incomplete Information},
author={Zeng, Pengxin and Yang, Mouxing and Lu, Yiding and Zhang, Changqing and Hu, Peng and Peng, Xi},
journal={arXiv preprint arXiv:2305.12743},
year={2023}
}