What is the results value on opencv 3 random forest predict()

asked 2018-06-19 18:00:07 -0600

updated 2018-06-20 13:15:56 -0600

In python OpenCV 3+ the random forest forest documentation says predict() returns retval and results. It is clear that retval is the predicted label, but results is simply defined as "The optional output matrix of results".

For example, here is my numpy array that I pass in as a single input:

array([176.20, 39.10, 116.55,  0., 0., 48.22, 76.83, 0.], dtype=float32)

These are the values I get for results:

array([[1.],
       [1.],
       [4.],
       [1.],
       [1.],
       [1.],
       [4.],
       [3.]], dtype=float32)

retval is:

1.0

What is the format and/or meaning of the array I get back?

edit retag flag offensive close merge delete

add a comment

1 answer

Sort by » oldest newest most voted

answered 2018-06-19 20:44:39 -0600

Tetragramm

7376 ●13 ●37

retval is just the first value of results. results is if your samples includes more than one input, so you can actually get all of them back.

edit flag offensive delete link

Comments

I've updated my answer to include an example of my results. I am passing in 1 value (numpy array with 8 elements). The first value of results corresponds to retval as you stated. However, I've only passed in 1 value so I'm not sure how that maps to the next 7 values in results. Could you clarify that a little more? I apologize if I'm a little slow. Thanks in advance!

Rick Smith ( 2018-06-20 13:20:37 -0600 )edit

I can't say anything without having seen how you train it. You're getting one response per element, so maybe it thinks each of these is an input value.