Training Datasets

This is the collection of datasets used within the training courses delivered by intersect.

The links to download the latest version of the datasets are;

• intersect_training

Updating Datasets

There are two sets of releases that will automatically take place. Upon updating the master branch, the Master release will be updated. The Master release is a pre-release and so doesn't modify the latest release, providing a method of testing changes.

The creation of a new release, which will update the data downloaded by participants can be done in one of two ways.

1. Using the git command line interface, create a tag git tag -a v0.1.7 and push the tag to GitHub ensuring the use of the --tags flag, git push --tags.
2. Using the GitHub web interface, create a new release setting the name of the new tag and the rest of the information.

Adding new datasets

New datasets can be added to the repository with some minor changes. This would involve creating a new directory alongside the intersect_training directory that contains the new dataset. With this dataset added, the next step is the automation, which is found in the files .github/workflows/draft.yml and .github/workflows/release.yml.

There are two changes that need to be made to each of these files. Firstly a new compression step needs to be added

      - name: Compress Directories
        uses: montudor/action-zip@v0.1.0
        with:
          args: zip -qq -r new_directory.zip ./new_directory

which has to come before the Create Release step. The final step is to add the name used for the zip file, here new_directory.zip to the end of the files key in the Create Release step giving

files: >
    intersect_training_data.zip
    new_directory.zip