by iterative.ai
Get started
Edit on GitHub

Get Started with CML on GitLab

Here, we'll walk through a tutorial to start using CML. For simplicity, we'll show the demo in GitLab CI/CD, but instructions are pretty similar for all the supported CI systems.

  1. Fork our example project repository.

    gitlab fork cml project

  2. ⚠️ Follow these instructions to configure a GitLab access token for CML.

The following steps can all be done in the GitLab browser interface. However, to follow along the commands, we recommend cloning your fork to your local workstation:

$ git clone https://gitlab.com/<your-username>/example_cml
$ cd example_cml

gitlab cml clone

  1. To create a CML workflow, copy the following into a new file named .gitlab-ci.yml:

    train-and-report:
      image: iterativeai/cml:0-dvc2-base1
      script:
        - pip install -r requirements.txt
        - python train.py
    
        - cat metrics.txt >> report.md
        - echo '![](./plot.png)' >> report.md
        - cml comment create --publish report.md
  2. In your text editor, open train.py and modify line 15 to depth = 5.

  3. Commit and push the changes:

    $ git checkout -b experiment
    $ git add . && git commit -m "modify forest depth"
    $ git push origin experiment
  4. In GitLab, create a Merge Request to compare the experiment branch to master.

    create merge request

    The "New Merge Request" page will let you Change branches:

    new merge request

    Ensure the target is your fork (under your username):

    change user name

    Continue and submit the Merge Request. Shortly, you should see a comment appear in the Merge Request with your CML report. This is a result of the cml comment create command in your workflow.

    cml start gitlab end

This is the gist of the CML workflow: when you push changes to your GitLab repository, the workflow in your .gitlab-ci.yml file gets run and a report generated.

CML commands let you display relevant results from the workflow, like model performance metrics and vizualizations, in GitLab comments. What kind of workflow you want to run, and want to put in your CML report, is up to you.

Final Solution

An example of what your repository should look like now can be found at iterative.ai/cml-base-case.

Content

🐛 Found an issue? Let us know! Or fix it:

Edit on GitHub

Have a question? Join our chat, we will help you:

Discord Chat