Thanks! I couldn't find that help!!
Simulating the decision of a bagged tree
3 vues (au cours des 30 derniers jours)
Afficher commentaires plus anciens
Elena Casiraghi
le 27 Fév 2020
Commenté : Elena Casiraghi
le 3 Mar 2020
Dear,
could anyone tell me how an object of a treeBagger takes a decision?
Precisely, if I have this code:
dataMat = randi(100, 200, 31) % toy example
labels = logical(randi(1,200,1)); % toy example
numTrees = 7;
mdlRF = TreeBagger(numTrees, dataMat,labels, 'CategoricalPredictors', strcmpi(vartypes, 'categorical'), ...
'PredictorNames', vars, 'PredictorSelection','interaction-curvature', ...
'Surrogate','on', ...
'Reproducible',true);
% new example
xx = randi(100, 1, 31);
pred = predict(mdlRF, xx);
How is pred computed?
The bagged tree has trained 7 trees which are saved in the cell array
mdlRF.Trees
I suppose that I should run a for loop such as this one:
for nL = 1: numTrees;
% each tree is used for predicting
pred(nL) = predict(mdlRF.Trees{nL}, xx)
end
% AFTER THE FOR LOOP VARIABLE pred contains the predictions of the numTrees classifiers
How are the different predictions composed to form the final prediction??
I used the average but it does not work.
Could anyone help me?
Réponse acceptée
Puru Kathuria
le 3 Mar 2020
Hi,
I understand that you want to know how predict function in TreeBagger takes its decision. The function predict returns a cell array of character vectors for classification and a numeric array for regression. By default, predict takes a democratic (nonweighted) average vote from all trees in the ensemble. For more information related to predict and the formula used to compute the predicted response, you can visit the following links:
0 commentaires
Plus de réponses (0)
Voir également
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!