Hi there - I wanted to clarify something about the test data.
I understand that we need to submit a CSV with the field IDs and a probability between 0 and 1, as per instructions. We have been given 11 months of data, with the same test shape file, so the same fields in all of them. I want to clarify whether we should be outputting a probabilities calculated for a particular month or the year, or if we can choose any month, or if we need to compute for all months and then from this produce one set of probabilites per field to submit? Thanks.
Or if we should have 11 rows for, say, field ID 5 ... one for each month.
I am guessing since the training data only shows one value for each farm, you are expected to find a way to only get one probability for each field. How you do that, is up to you.