name: Install test SINGE

# Builds and tests the SINGE Docker image, tests SINGE in a Docker container

docker:

name: Install test Docker

runs-on: ubuntu-latest

steps:

- name: Checkout repository

uses: actions/checkout@v2

# Pull from Docker Hub to use the cache

# https://medium.com/mobileforgood/coding-tips-patterns-for-continuous-integration-with-docker-on-travis-ci-9cedb8348a62

# https://github.com/docker/build-push-action/issues/7

- name: Pull Docker image

run: docker pull agitter/singe:tmp

- name: Build Docker image

uses: docker/build-push-action@v1

with:

username: ${{ secrets.DOCKER_USERNAME }}

password: ${{ secrets.DOCKER_PASSWORD }}

path: .

dockerfile: docker/Dockerfile

repository: agitter/singe

tags: tmp

cache_froms: agitter/singe:tmp

add_git_labels: true

# Only push when running on the master branch

push: ${{ github.ref == 'refs/heads/master' }}

- name: Test SINGE

run: |

# Installs MATLAB runtime R2020a and tests compiled SINGE for macOS on test data

macos:

name: Install test macOS

[](https://github.com/gitter-lab/SINGE/actions?query=workflow%3A%22Install+test+SINGE%22)

The [`default_hyperparameters.txt`](default_hyperparameters.txt) corresponds to an ensemble obtained using **ten** subsampled replicates for each hyperparameter combination.

However, as noted in the manuscript, the user can substantially reduce the computational runtime by using **two** to **five** replicates instead of **ten** at the cost of moderate precision-recall performance degradation.

This strategy combined with using a regulator list as described in the following section can drastically reduce the computational requirements for large datasets.

See the [instructions for generating a hyperparameter file](scripts/README.md) for further details.