A project collaborator can click add script to either upload a Python or R script to the project, or create one in the Watson Studio Local text editor. You can also generate and save a script into your project when you run a batch score or evaluate of a machine learning model.
Create a script
To create the script in the Watson Studio Local text editor, go to the project assets and click add script. You can also click on existing scripts to edit them.
Tasks you can perform in the script editor:
- Save your script in the project.
- Run the script using an automatically created job. This same job is run on each ensuing click of the run button.
- Specify environment variables and command line arguments for each run.
- View a history of the script runs. Click any run ID to view its logs.
- Configure settings in your text editor, including Auto-completion and an Auto-save function that automatically saves your progress every minute.
Test a script
To test a script in a development environment with Python 2, Python 3, or R, click Test script as API next to it. Watson Studio Local automatically generates curl commands to retrieve a bearer token and run the script. You can copy and paste the commands and replace the variables accordingly.
Run a script
To run a script, click Create job next to it and schedule a job for it.