36 Building Your Own Gesture Recognition System with Your Own Data

[@sharanyabhat8473]

Hello, I am using 5000+ images for training and 1000+ for testing.. I am getting validation accuracy 0 for this code and also real time it doesn't predict at all...

[@nailyakhafaeva1298]

hi! were you able to solve the problem?

[@SunnySingh-ug3xy]

How to add more numbers like 2,3,4,5 upto 9???

[@exterminar7230]

mega.nz/file/aFIghBaS#UFeJmmsjyTraJ6lnPrQYKHcGUMMBnTash-K4r_3GdqM

[@SunnySingh-ug3xy]

I have already downloaded your code but I want to train 0-9 and all alphabets... This code is working only for binary..
please tell me how to do it..

[@sahilshirodkar6940]

@@SunnySingh-ug3xy where u have downloaded

[@ramachandrabaddireddy3204]

How to train remaining numbers from 2 to 9

[@univip2366]

Sir,how do i train all 26 letters using this... please help sir

[@tonynguyen2201]

I got some errors. Because we can not use predict_classes anymore so i changed it into np.argmax(classi.predict(roi, 1 , verbose = 0), axis=-1). But when I check with camera, it does not return any other value instead of 0. Just stay at [0] and nothing else happens even though i try alot of times. Please help

[@ipatimo]

for me it works like this:
result = classifier.predict(roi)
number = np.argmax(result)
cv2.putText(copy, str(number), (300 , 100), cv2.FONT_HERSHEY_COMPLEX, 2, (0, 255, 0), 2)

[@debbyverseyt6666]

@@ipatimo hello please i got an error that classifier is not defined

[@debbyverseyt6666]

how do i fix this please?

[@ipatimo]

@@debbyverseyt6666 classifier is your model. What is it name? Or you need to run the script from the beginning, because the kernel has crashed and forgot everything.

[@debbyverseyt6666]

@@ipatimo thank you i've gotten to work but now my frame is too small