Skip to content

Variational Representation Learning by using proximity loss functions

Notifications You must be signed in to change notification settings

vikigenius/prox_vae

Repository files navigation

Proximity VAE

Variational Representation Learning by using proximity loss functions

The data and saved models can be downloaded from: resources

Replicate the conda environment in the environment file using: conda env create -f environment.yml

Train the model using the following command:

allennlp train -s models/info_vae_snli_cosine_reg --include-package src \
    config/info_Vae_snli_cosine_reg.jsonnet

Can perform transfer using the following command:

python prox_vae.py transfer models/info_vae_snli_cosine_reg --templates_file \
    data/interim/snli_1.0/templates.tsv

About

Variational Representation Learning by using proximity loss functions

Resources

Stars

Watchers

Forks