Is the value in the simplified_snapshot column in rca_prior.csv the alarm_id or the device_id?
In dataset_1, the maximum value of simplified_snapshot is 34.
In dataset_3, the maximum value of simplified_snapshot is 38.
Both datasets match the device_id range.
However, the description in rca_prior.csv makes it look like alarm_id.
https://gcastle-hub.github.io/competions.github.io/assets/img/rca_prior.png