Define standard: predictions

[terryf82]

We need to define an initial standard for predictions the project generates, that resolves:

• required / optional attributes of a prediction, including data typing

• temporal aspect: previously we've talked about converting the dates for weekly predictions from a 1-53 week identifier to an epoch timestamp (e.g. the first second of the week) to ensure that we build with the capability to handle more than one year

[terryf82]

@bpben is this the right place to talk about outputting data on how a prediction is reached?

I think one fundamental aspect of generating predictions that cities would be willing to action is being able to explain how they were reached (e.g contributing factors and their relative weight). What are your thoughts on the best way to output this?

@j-t-t

[bpben]

Ahh, good point. I think I'd like to standardize the pipeline first. Then we can see how data is coming in and adjust accordingly.

[terryf82]

@bpben based on this morning's meeting it seems like we're pushing this back to a future release (possibly v2) to allow us to clear up some more pressing jobs first?

[terryf82]

@bpben this one has been around for a long time now, do we want to have a go at defining a standards yet or still not enough value?

[bpben]

I do want to take this up, will add to the map - should be part of revamping the model pipeline