Could someone remind me how to propagate changes across Partitioned assets? Here I have some

data

from dagster import StaticPartitionsDefinition, asset, FreshnessPolicy

oceans_partitions_def = StaticPartitionsDefinition(
    ["arctic", "atlantic", "indian", "pacific", "southern"]
)

@asset(freshness_policy=FreshnessPolicy(cron_schedule='0 */1* * *', maximum_lag_minutes=5)) # Update this every hour
def data()->pd.DataFrame:
    return pd.read_sql('SELECT * FROM my_ocean_data_that_updates_every_minute')

@asset(partitions_def=oceans_partitions_defs) # How do I mark this as stale when `data` updates?
def ml_model_for_each_ocean(context, data):
    relevant_data = data.loc[lambda x: x['ocean'] == context.asset_partition_key_for_output()]
    model = build_model(relevant_data)
    return model

Hi Mycchaka. Do you mean that the downstream partitioned assets are not being marked as stale in the UI? Unfortunately, we don't yet support displaying staleness for partitioned assets. Here's an issue for tracking: https://github.com/dagster-io/dagster/issues/12314