Skip to content

How do you run the baseband polarization pipeline on an event?

Similar to the baseband localization pipeline, you just need to add an event number to the bucket: baseband-polarization which can be done from anywhere you have your FRB Master access and refresh tokens (e.g., opening up IPython on frb-analysis):

Python
In [1]: import chime_frb_api

In [2]: buck = chime_frb_api.bucket.Bucket(base_url="https://frb.chimenet.ca/maestro/buckets")

In [3]: work = {"event_number": 195718715, "run_stage": "polarization"}

In [4]: buck.deposit('baseband-pipeline', work=work, priority="low")
Out[4]: [True]

The default priority for an event is low, but can also be medium or high. In this case, events with high priority will be analyzed first after finishing the current event.

N.b. The pipeline can only be run on events with a preexisting singlebeam_<EVENT_ID>.h5 file. If one does not exist, you must first run the Localization Pipeline.

Modifying parameters

There are no optional parameters at this time. This feature will be incorporated in a future release.

Generated files

The polarization pipeline generates pol_prod.json file that summarizes the results as well as plots from various stages of the pipeline. These outpuut files can be found within the plots/singlebeam subdirectory.

Troubleshooting

The pipeline's performance on an event can be monitored in realtime by consulting the analysis cluster monitoring page. Otherwise, logs can be retrieved after the event's completion (or error) and found in the logs subdirectory. The error logs corresponding to the polarization pipeline is baseband-<EVENT_ID>-polarization.log.