This repository contains the code and supporting files for the ASReview simulation project. It is primarily intended for use with the paper (provide paper reference here) and offers a comprehensive package for running ASReview simulations, launching the cluster, and analyzing the results.
The project is composed of three main components:
- The ASReview-simulate Docker image
- Exoscale cluster launch scripts
- Output analysis scripts
The Docker image provides a pre-configured environment for running the ASReview simulation tool. The image is based on the official Python 3 Docker image and includes all dependencies required for running ASReview simulations.
The image uses the Synergy dataset, and will run the ARFI template for a single dataset using Makita.
To pull the image from GitHub Container Registry, run the following command:
docker pull ghcr.io/jteijema/asreview-simulation-project:latest
docker tag ghcr.io/jteijema/asreview-simulation-project:latest asreview-simulate
To use the image, run it with docker run
, passing any required environment variables as -e
arguments. For example:
docker run --rm \
-e EXOSCALE_API_KEY=<access_key> \
-e EXOSCALE_API_SECRET=<secret_key> \
-e EXOSCALE_STORAGE_ZONE=de-fra-1 \
-e BUCKET=<bucket_name> \
-e DATASET=<dataset> \
-e SETTINGS="<settings>" \
asreview-simulate
Replace <variable> with the appropriate values.
The following environment variables are required:
EXOSCALE_API_KEY
: the Exoscale access key IDEXOSCALE_API_SECRET
: the Exoscale secret access keyEXOSCALE_STORAGE_ZONE
: the Exoscale storage zone (e.g., de-fra-1)BUCKET
: the name of the Exoscale bucket to useDATASET
: the name of the dataset to use from synergySETTINGS
: the settings for the simulation
You can build the image yourself by cloning this repository and running the following command:
docker build -t asreview-simulate:latest .
The exoscale
folder contains scripts for launching the cluster on Exoscale, and generating jobs for this cluster.
The output
folder contains scripts for analyzing the results of a simulation study.
Here's an example command that runs a simulation with the provided environment variables:
docker run --rm \
-e EXOSCALE_API_KEY=<access_key> \
-e EXOSCALE_API_SECRET=<secret_key> \
-e EXOSCALE_STORAGE_ZONE=de-fra-1 \
-e BUCKET=asreview-output \
-e DATASET=Donners_2021 \
-e SETTINGS="-m nb -e tfidf" \
asreview-simulate
Replace <access_key> and <secret_key> with your Exoscale access key ID, secret access key, and account name, respectively.
An example of the results of the simulation can be found on the simulation website.. For these results, the docker image was ran 338 times.
This project is licensed under the terms of the Attribution 4.0 International (CC BY 4.0). See the LICENSE.md file for details.