Get Started with CML on Bitbucket

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

1. Fork our example project repository.

   

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

3. ⚠️ Follow these instructions to enable the Pull Request Commit Links application.

$ git clone https://bitbucket.org/<your-username>/example-cml

4. To create a CML workflow, copy the following into a new file named bitbucket-pipelines.yml on your master branch:

   image: iterativeai/cml:0-dvc2-base1
   pipelines:
     default:
       - step:
           name: Train model
           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

5. In your text editor, open train.py and modify line 15 to depth = 5.

6. Commit and push the changes:

   $ git checkout -b experiment
   $ git add . && git commit -m "modify forest depth"
   $ git push origin experiment

7. In Bitbucket, create a Pull Request to compare the experiment branch to master.

   

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

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

   

This is the gist of the CML workflow: when you push changes to your Bitbucket repository, the workflow in your bitbucket-pipelines.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 Bitbucket checks and 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.