/visual-cnn-lrp

Added REST interface and explanation method to http://scs.ryerson.ca/~aharley/vis/

Primary LanguageJavaScript

Visual CNN LRP

Add REST interface and layer-wise relevance propagation method to http://scs.ryerson.ca/~aharley/vis/

Installation

For the backend install Python 3 and Flask and then from the backend directory start the Flask server by entering FLASK_APP=backend.py flask run. Currently the frontend expects the backend to be on port 5000. For the frontend you should be able to use any web server, I chose nginx. Point the webserver at the webui folder and start it.

Run Docker Container

In order to run the visualization in docker containers (one for frontend and one for backend) you can execute the docker_build.sh and docker_start.sh files.