Taxi demand predictor service (WIP)

Summary

Use case: we are working for a NYC Taxi company, and we would like to optimize the number of cars available at any time. This means having low number of cars available at low-demand hours, and many more at peak hours. As a simple exercise, we would like to make a prediction on the demand of rides at any hour.

Project structure

The notebook structure is as follow:

• 01_get_data:

• 02_transform_raw_data_into_ts:

• 03a_feature_eng_1:

• 03b_feature_eng_2:

• 04_traning: