Output confusion matrix in Weka from command line -
I've saved using Weka 3.7.9 a random forest model file, and now I try to evaluate it I am against the other (very large) set (on some large machines in Amazon EC2) I am using the following command line:
& gt; Java -server -Xmx60000m -cp weka.jar weka.classifiers.Evaluation weka.classifiers.trees.RandomForest -T test.arff -l random forest.model I -No-CV However, only production I mean something like this:
=== test data === correctly classified examples 3252532 80.0686% by mistakenly classified examples 809,651 19.9314% Kappa error statistics 0.2884 full error 0.2539 root error means 0.3608 cases (0.95 level) coverage of the number of 98.7413% of the total paid-up examples 4,062,183 while I anything except I'm looking for a way:
=== Extended purity class === T.P. Rate FP rate Precision Recall F-measure MCC ROC area PRC area Class 0.804 0.295 0.731 0.804 0.766 0.512 0.826 0.803 buyers 0.705 0.196 0.783 0.705 0.742 0.512 0.826 0.798 Non-buyer to the weighted average. 0.755 0.245 0.757 0.755 0.754 0.512 0.826 0.801 === Confusion Matrix === One B & L; - Categorized as - 61728 15004. A = Buyer 22662 54066 | B = non-buyer Please note that, even though I run entirely training method fully would, like:
& gt; Java-Xmx60000m -cp weka.jar weka.classifiers.Evaluation weka.classifiers.trees.RandomForest income tax train.arff -t test.arff I 10KS1 -num-slot 8 random-one -d Model -i -no-cv I still do not show confusion metrics (only for trained data) for exam data.
"itemprop =" text "> This works when you leave the no-cv option .
Comments
Post a Comment