small fix: removing repartition as it has no effect before groupby

[ryuta-yoshimatsu]

Removing repartition from the following code snippet:

before

score_one_model_fn = functools.partial(Forecaster.score_one_model, model=model)
res_sdf = (
    src_df.repartition(100)
    .groupby(self.conf["group_id"])
    .applyInPandas(score_one_model_fn, schema=output_schema)
)

after

score_one_model_fn = functools.partial(Forecaster.score_one_model, model=model)
res_sdf = (
    src_df.groupby(self.conf["group_id"])
    .applyInPandas(score_one_model_fn, schema=output_schema)
)

as it has no effect preceding the groupby.

[ryuta-yoshimatsu]

Hi @dbbnicole, it seems like the DEPLOYMENT_TARGET_TOKEN_GCP has expired and we are getting the following error when running CI on PR:

Error: {"error_code":"403","message":"Invalid access token."} at IncomingMessage.<anonymous> (/home/runner/work/_actions/databricks/run-notebook/v0/dist/webpack:/typescript-action/lib/common/src/request.js:46:1) at IncomingMessage.emit (node:events:525:35) at endReadableNT (node:internal/streams/readable:1358:12) at processTicksAndRejections (node:internal/process/task_queues:83:21)

Should we generate a new token using a service principal and set it up in GitHub?