feat: Add ookla processing module

[tm-jc-nacpil]

Overview

Adds a new osm.py module that provides:

• add_ookla_features() one-liner to add Ookla features to an AOI. Features can be specified by year and type (fixed, mobile)
• OoklaDataManager convenience class to load and automatically cache Ookla data on disk and RAM. It caches the processed data per aoi/type/year combination

Addresses #67

Notes

• In add_ookla_features, it uses all of the available features in the raw dataset (hardcoded). I want to incorporate something like the geowrangler agg spec so that we can select some features and process/rename them independently
• So far, OoklaDataManager only caches the raw files. Next step would be to cache also the processed data to the filesystem (identified by the hash)

Starting point

test-notebooks/test_refactors/test_ookla.ipynb: test notebook to demo the data manager and feature engineering modules. It currently uses phl_dhs_cluster_level.csv

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB