[Artificial Intelligence in practice] Faces recognition: SVM classification model – part 1


In face recognition problem I’ve used preprocessed data set of the “Labeled Faces in the Wild” (http://vis-www.cs.umass.edu/lfw/lfw-funneled.tgz).
Our aim is expected results for the top 5 most represented people in the dataset and the algorithm should achieve convergence (values in support column should be the same or close to expected value)

Total dataset size:
Number of samples faces to recognize: 1288
Number of features: 1850
Number of classes: 7

Extracting the top 150 eigenfaces from 966 faces: done in 0.443s
Fitting the classifier to the training set:  done in 21.080s
Predicting people’s names on the test set:  done in 0.043s


In our example algorithm works PERFECT and has achieved convergence because values from support columns are the same

soon will be more info…