VaLPACa

This repository contains code to run the VaLPACa algorithm (Variational Ladders for Parallel Autoencoding of Calcium imaging data), as published in Prince et al., 2021, biorXiv.

The code is written in Python 3.
Requirements for running the scripts in this repository are listed in requirements.txt.

To train VaLPACa on a dataset, run train_model.py.
To extract latents from data, using a trained model, run infer_latent.py.

To train VaLPACa on a dataset, run train_model.py.
To extract latents from data, using a trained model, run infer_latent.py.

• analysis: latent analysis modules
• batch: example slurm scripts
• data: data generation and processing modules
• hyperparameters: yaml files defining hyperparameter values for different models and datasets
• models: modules defining models, and objective functions
• utils: processing, training, and plotting utilities

Follow the link below for an example of how to run VaLPACa on new data.