insight-lane / crash-model

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.
https://insightlane.org
MIT License
113 stars 40 forks source link

Add transit stops as a feature #190

Closed j-t-t closed 1 year ago

j-t-t commented 6 years ago

Prioritized because it's been requested by Philadelphia stakeholders

nsteins commented 4 years ago

Any thoughts or progress on this? I tried adding bus stops via the supplemental csv pattern, but the stops don't have a date associated with them, which I think breaks the functionality. I think we could use the standardize_point_data.py script if we are able to skip the date aspect. I just don't know what that would break down the line.

j-t-t commented 4 years ago

I think the goal was to use a standardized data source so that adding transit stops wouldn't need to be a one-off for each city. Open Street Maps has some of this data, and there are other transit info data sources that we were hoping to use.

You can certainly add your transit stops as a one off (but I wouldn't close this task if you do) via point-based features. If I recall correctly, the reason date was required was so that if a particular feature didn't exist during a window of dates when you were looking at crashes, it would get filtered out. If you make the changes so that date is not required, and then to test:

If I'm not recalling this correctly, let me know and I can dig into it. It would be great to add these checks as unit tests as well. Let me know if I can help.

bpben commented 1 year ago

318 should have unlocked this! @nsteins - I know this is a blast from the past but if you're interested in contributing here, let me no. I'm going to close this as it's city-specific and we might need a fresh request to reopen.