Schedule recurring pipeline runs
Recurring runs are cron-style triggers that create new pipeline runs on a schedule. Use them for nightly retraining, periodic data ingestion, or regular model evaluation.
Set up cron-style schedules to trigger pipeline runs automatically.
Prerequisites
- An uploaded pipeline (see Create a pipeline)
- The Vantage SDK installed and configured
Create a recurring run
Use the Vantage SDK to set up a schedule:
from vantage_sdk import VantageClient
client = VantageClient()
schedule = client.pipelines.create_recurring_run(
pipeline_name="training-pipeline",
name="nightly-retrain",
cron="0 2 * * *", # 2 AM daily
params={"data_path": "/data/raw/latest.csv"},
)
print(f"Schedule created: {schedule.id}")
Success looks like this: the schedule is created, and the pipeline will execute automatically at the configured intervals.
Common cron patterns:
| Pattern | Schedule |
|---|---|
0 2 * * * | Daily at 2:00 AM |
0 */6 * * * | Every 6 hours |
0 0 * * 1 | Weekly on Monday at midnight |
0 0 1 * * | Monthly on the 1st at midnight |
Manage recurring runs in the UI
Open the Pipelines page
Click Workbench in the left sidebar, then click Pipelines.
Navigate to Recurring Runs
Click the pipeline name, then select the Recurring Runs tab.
Browse the schedule list
The list shows:
| Column | Description |
|---|---|
| Name | Schedule display name |
| Pipeline | Pipeline version triggered |
| Cron schedule | The interval expression |
| Status | Enabled or Disabled |
| Last run | When the schedule last triggered |
| Next run | When the next trigger will fire |
Enable or disable a schedule
Click the row's action menu to Enable or Disable the schedule. Disabling pauses future triggers without deleting the configuration.
Delete a recurring run
Click the action menu and select Delete. Runs already triggered by the schedule are not affected.