The easiest way to get started is to clone the repository locally. If paperboy.db exists, simply delete it, this is my debug database. In the Makefile you'll see the runsql target. This runs with sql authentication, and a sqlite backend. Run it by running python3 -m paperboy.server --backend='sqla' --auth='sqla'
Navigate to port 8080. At the login screen, simply click Register and choose any username. There is no password with dummy auth.
Navigate to the Notebooks tab, then the uploader, and upload a notebook of your choice (I recommend examples/sample.ipynb included in this repo.
From there, you can start to setup jobs, which will dump Airflow compatible DAGs into ~/airflow/dags by default.
Navigate to the Jobs tab to add your first job. As an example, we will configure the sample.ipynb notebook to run with the papermill parameters defined in examples/params_min.jsonl.
Since the autogen reports checkbox was full, we autogenerate 6 reports corresonding to the 6 different parameters in the file. As a result, you should now see these in the reports tab:
You should also see a new dag in ~/airflow/dags, and the dag in the airflow UI. If airflow is configured to start dags automatically, you will soon see 6 reports in ~/Downloads every minute until you halt the job.
