UNAPEI/CapFALC NLP Docker Image

Purpose

The purpose of this project is to setup easily an environment to run CapFalcNLP toolbox for the UNAPEI application : CapFalc application.

This application should be able to run on a dedicated instance of the SSPCloud datalab application Onyxia to demonstrate the possibilities of the platform.

Docker Image

The image is publicly available on DockerHub.

Pre-requisite

To build this image you'll need at least to have git and docker installed on a linux box.

Config

A config directory contains the properties used by the scripts to build and run the image.

Here are the few necessary properties :

• name : the name of the docker image, by default : cleyrop/codatalab-capfalcnlp
• version : the version of the docker image, e.g. 0.0.1
• container : the name of the running container when run with default script, e.g. capfalcnlp

Scripts

The project contains a scripts directory with some easy to remind scripts/commands :

Build

To build this image, once you've clone the code, just run :

./build.sh [--no-cache]

[--no-cache] option can be passed to force build without cache.

Test

If you want to test the image content, from the scripts directory run :

./check.sh

To get some informations about the content of the image, libraries, extensions and versions, from the scripts directories, run :

./check_versions.sh

Run

To run this image localy, from the scripts directory, run :

./run.sh

Deploy

Deployment to dockerhub can be done, by running the following command from the scripts directory :

./deploy.sh

ALL

To run all commands in a raw, use :

./all.sh [--no-cache]

Exec

To execute a shell in the docker container (as root by default but could be changed)

./exec.sh [--no-cache]