Skip to content

LumionHXJ/GraphPL

Repository files navigation

GraphPL: Leveraging GNN for Efficient and Robust Modalities Imputation in Patchwork Learning

This repository contains code for the ICASSP'26 paper: GraphPL: Leveraging GNN for Efficient and Robust Modalities Imputation in Patchwork Learning. Our repository is built on CLAP.

introduction for files

  • mmnist: dataset, networks and running files for the PolyMNIST dataset.
  • mnistsvhntext: dataset, networks and running files for the MNIST-SVHN-TEXT dataset.
  • celeba: dataset, networks and running files for the CelebA dataset.
  • muse_eicu: dataset, networks and running files for the eICU dataset.
  • modalities: modality interface of the datasets.
  • eval_metrics: evaluation of the model, including reconstruction MSE, generation coherence, log-likelihood and latent space classification.
  • run_script/run_epochs_vaes.py: training script of all methods for all the datasets.
  • utils: needed function for the implementation.

requirements

the needed libraries are in requirements.txt.

dataset preparation:

The preparation of the PolyMNIST and MNIST-SVHN-TEXT datasets follows MoPoE.

Quad-CelebA

Based on the MoPoE, the images of CelebA is processed to extract edges using the Canny algorithm, and the face segmentation map is generated by the Face Parsing Model trained on CelebM from FaRL.

eICU

eICU data set is not directly available and detailed description is in the paper 'Benchmarking machine learning models on eICU critical care dataset'.

After obtaining the eICU data through application, it can be processed using MUSE, and then the most frequently occurring parts from diagnosis, treatment, and medication are selected until the cumulative probability reaches 90%.

experiments

All launch scripts are located in the scripts folder, where run_{dataset_name}.sh corresponds to CLAP and run_{dataset_name}_graph.sh corresponds to GRAPHPL. To reproduce other methods, you need to modify run_{dataset_name}.sh to the corresponding impute_method (poe for MVAE, moe for MMVAE, joint_elbo for MoPoE-VAE) and set k_single to 0.

Contact

If you have any questions, please contact huxingjian@pku.edu.cn.

Acknoledgements

We would like to thank the authors of MoPoE, MUSE, FaRL, and CLAP for their valuable contributions.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors