DRAVA: Utilizing Disentangled Representation Learning as A Visual Analytics Method for Pattern-based Data Exploration
Check out our preprint at https://osf.io/gzu27/
This repository has two main components: a frontend interface and a back-end server.
Please refer to the documentation webpage for more details
docsify serve docs
The backend is developed and tested with python@3.7.9
First, install all dependent packages:
cd server
pip install -r requirements.txt
Then, start the flask server:
cd flask_server
python app.py
To manage dependencies more effectively, you can create and use a conda environment before installing all packages:
conda create -n drava
conda activate drava
# ...
conda deactivate
The pre-trained models are stored at server/flask_server/saved_models
.
First, install all dependent packages:
cd front
npm install
Then, launch the Drava react application on the browser:
npm start
To run both the server and the client, you need to put additional files into your local repository. These include
server/data/
(Image patches of JPG files and compressed numpy arrays of.npz
files)front/src/assets/
(JSON files that specify genomic ranges)front/public/assets/
(CSV files that contain external analysis results)
These datasets are shared upon request.