Build a crash prediction modeling application that leverages multiple data sources to generate a set of dynamic predictions we can use to identify potential trouble spots and direct timely safety interventions.
There are several different interpretations of the _startyear and _endyear values provided in the city-specific configs:
in some situations, end_year is treated as 01-01-end_year and data is filtered out if it it occurred in the same year as end_year (that is, if end_year is 2018 and a crash happened in 2018, the crash is filtered)
in other situations, end_year is treated as 31-12-end_year and data is filtered out only if it occurred in a year after end_year (so if end_year is 2018 crashes in 2018 are included, crashes in 2019 are excluded).
We need to review all uses of these values and align under a common understanding. It's probably worth swapping to full start_date and end_date values that include day, month and year to clear up any confusion.
There are several different interpretations of the _startyear and _endyear values provided in the city-specific configs:
in some situations, end_year is treated as 01-01-end_year and data is filtered out if it it occurred in the same year as end_year (that is, if end_year is 2018 and a crash happened in 2018, the crash is filtered)
in other situations, end_year is treated as 31-12-end_year and data is filtered out only if it occurred in a year after end_year (so if end_year is 2018 crashes in 2018 are included, crashes in 2019 are excluded).
We need to review all uses of these values and align under a common understanding. It's probably worth swapping to full start_date and end_date values that include day, month and year to clear up any confusion.