Error occurred when I used tensor flow lite custom model

sng
1

I faced an issue while running an object detection program using tensor flow lite costum model.I am attaching the error code bellow.

classes = interpreter.get_tensor(output_details[1]['index'])[0] # Class index of detected objects IndexError: list index out of range

Comments

off topic. get_tensor is not an opencv function?

LBerger ( 2020-03-05 06:43:19 -0600 )edit

and unfortunately, so far there is no support for tf lite models in opencv's dnn module

berak ( 2020-03-05 07:00:06 -0600 )edit

supra56 ( 2020-03-05 07:29:49 -0600 )edit

943 ●9 ●6

Change this:

classes = interpreter.get_tensor(output_details[1]['index'])[0]
to:
classes = interpreter.get_tensor(output_details[1]['index'])

Are you looping after?

supra56 ( 2020-03-05 07:26:55 -0600 )edit

we are getting the same error. This is working with the existing model but not with the customised model. Kindly reply.

sng ( 2020-03-06 22:29:07 -0600 )edit

1 follower

Asked: 2020-03-05 06:28:08 -0600

Seen: 790 times

Last updated: Mar 05 '20