In parameters.json, company_id contains all integers from 0 to 100 while in ground_truth.csv, company_id only contains integers from 0 to 56. May I ask if it is a mistake or the test data may contain values larger than 56?
payment_type contains all integers from -1 to 8 except 4. Does payment_type of the test data contain 4? Thanks.
So, it turns out that the rise of Uber and Lyft has been rough for the taxi industry. The number of cab companies (different values for company_id) does vary noticeably between different data sets. The company_id range is legitimately from 0-100 (and the note that the same company may be assigned a different code in different data sets).
However, payment_type 4 does not appear to be used in any of the data sets. You’re welcome to omit that option.