Skip to content
/ MGVAE Public

Code for "Generative Oversampling for Imbalanced Data via Majority-Guided VAE", AISTATS2023.

Notifications You must be signed in to change notification settings

Aiqz/MGVAE

Repository files navigation

MGVAE

Code for "Generative Oversampling for Imbalanced Data via Majority-Guided VAE", AISTATS 2023.

Usage

For MGVAE training:

python mgvae_train.py -c "your config file!"

Note that all config template of generative models used in our experiments are in "configs/configs_generation" directory, you need to add your data path for successfully run.

For classifier trainig:

python classifier_train.py -c "your config file!"

Note that all config template of classifiers used in our experiments are in "configs/configs_DATASET" directory.

About

Code for "Generative Oversampling for Imbalanced Data via Majority-Guided VAE", AISTATS2023.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages