Progress bar for training

FaceRecognizer

asked 2017-01-02 03:36:33 -0600

Hi everyone

I am experimenting with the Fisher Recognition face module and using Python on a Raspberry.

At the part model = cv2.createFisherFaceRecognizer() model.train(images, lables)

it just waits there and there is no way to know how long would it take. Yes, I know a raspberry is not a superserver so it will take time even with just a few images, but having some kind of progress bar would be nice.

I tried to google for one but to no avail, so I am obliged to recur to the community's good will to help me find a progress bar to the training part of the Fisher Face Recognizer system, and to that I thank beforehand.

edit retag flag offensive close merge delete

add a comment

1 answer

Sort by » oldest newest most voted

answered 2017-01-02 04:04:46 -0600

berak
32993 ●7 ●81 ●312

updated 2017-01-02 05:12:46 -0600

though i often wanted to have something like this myself, i don't think, this is possible at all.

the training code spends 99% of the time in the PCA, and you cannot insert some kind of callback there.

imho, all you can do , is make a few measurements (say, 50, 100, 300 images), and extrapolate from there. (it grows exponential with number/size of images)