Creating query parameters for "intra-city" run

[ChristinaLast]

• [x] adding cities-option to cohort builder and time splitter
• [x] filtering for reference grade sensors.

[ChristinaLast]

This branch is currently hard coded for Cairo, EG. You can change those directly to suit your usecase, or you can fix the code to produce --city or --country filters. I had the challenge of making then work together and error handling if one returns None within a specified time bound

[ChristinaLast]

I am trying to build validation set with

https://api.openaq.org/v2/measurements?date_from=2023-07-07T00:00:00Z&date_to=2023-10-07T00:00:00Z&parameter_id=2&country=US

but there is an issue with using the parameter or the parameter_id flag to get pm2.5 data only. response is:

{
    "detail": "Connection timed out"
}

[ChristinaLast]

There is some issues with doing only 3 month time window, I am pushing 12 months to branch as this is working.

[ChristinaLast]

End to end pipeline is working on this branch, error handling performed in the following ways:

• I am using try: except to deal with the errors in imputation (e.g. not enough data values in a column leading to no capacity to impute values)
• I am using try: except to handle cases where due to no imputation prediction is not able to be performed on the test set.