Repository navigation
Testing Works Great, How about new Predictions? #6
Description
I'm going to continue to dive into the code, but I can't seem to figure out how if I have a trained forest off a data set, example:
[[1, 2, 3, 4]
[5, 6, 7, 8]
[9, 10, 11, 12]]
how to get RandomForest to predict the fourth (last) position from the test dataset, example:
[[13, 14, 15, ___ ]]
Of course, this isn't my application-- consecutive numbers, but it is the easiest example I can provide.
I haven't seen yet anything provided in the examples or documentation what happens after Run().
As I've said what I'm specifically asking for is a way for RandomTree to provide me it's vote on the last position of a data set if it can't do that already.. thanks!
Activity
ironmanMA commentedon Jul 2, 2018
as mentioned #2 you could see how to train a dataset and then run on new data.
if i understand what you're really asking for here is a way to understand vote distribution of all the trees in the forest.
now thats really interesting proposition @evanschoffstall .