Getting the probability values in the random forest classification template of PredictionIO

[rbag...@gmail.com]

Hey,

We are using the classification template of PredictionIO (algorithm modified to random forest instead of Naive Bayes'). We wanted to ask how is the random forest classifying a test input: By the number of votes each tree give or the mean of the probability of each label in each tree)? Also, can we extract the mean probability of each class for a particular input so that we can get an idea of how is the strength of the prediction? And does this probability value reflect the absolute tendency of the input to belong to the said class?

[Pat Ferrel]

You probably should ask on PredictionIO mailing list.

Signup here: http://predictionio.incubator.apache.org/support/

On Jun 29, 2017, at 10:33 PM, rbag...@gmail.com wrote:

Hey,

We are using the classification template of PredictionIO (algorithm modified to random forest instead of Naive Bayes'). We wanted to ask how is the random forest classifying a test input: By the number of votes each tree give or the mean of the probability of each label in each tree)? Also, can we extract the mean probability of each class for a particular input so that we can get an idea of how is the strength of the prediction? And does this probability value reflect the absolute tendency of the input to belong to the said class?

--
You received this message because you are subscribed to the Google Groups "actionml-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to actionml-use...@googlegroups.com.
To post to this group, send email to action...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/actionml-user/e1e2291a-bdb6-4dc4-8d27-73ef1f1b57f8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.