Create cloud function to save county CSV to bucket

[isaacmg]

• Deploy cloud function on GCS and test on incoming data.

[efawe]

Do we have a county csv already or do we need to aggregate one ?

[isaacmg]

So @efawe the main thing for this is to just create a cloud function to split up the data that you persisted in #48. This is currently already done in notebooks but in order to clean up notebooks and have unified county data having separate csv for each county would be useful. Also when we go to deploy the models to production in #60 we will need data persisted on a county level.

[isaacmg]

So just modify this function to output a CSV file in the format task_ts/date/county_name.csv to the proper path.

task-ts/corona_ts/data_utils/data_creator.py

Line 4 in 23a77bf

def loop_through_locations(

I've previously developed cloud functions so I might be able to give you some additional code. Just create the cloud function on Google then have it be triggered on finalize.

[efawe]

Figured out cloud functions. Should have the full code in a day

[efawe]

Created the cloud function. How should I add some form of copy on the github?

[isaacmg]

@efawe Is it working on GCP? If so open a PR containing the code and the requirements.txt file. Store it a directory in the following way cloud_functions/partition_county/main.py and cloud_functions/partition_county/requirements.txt Thanks.

[efawe]

Yes and will do

[efawe]

Created a separate issue that is more reflective of the development. Currently google cloud functions is not practical financially.
#69 #