While using detection example with ssd mobilenet v2 everything works fine, when i load ssdlite mobiledet and using int8 quantized model, i get the following error:
Traceback (most recent call last):
File "detect_image.py", line 130, in <module>
main()
File "detect_image.py", line 109, in main
objs = detect.get_output(interpreter, args.threshold, scale)
File "/home/pi/coral/tflite/python/examples/detection/detect.py", line 153, in get_output
count = int(output_tensor(interpreter, 3))
TypeError: only size-1 arrays can be converted to Python scalars
On the other hand, when using fp32, the example code is running but the detection result is false:
----INFERENCE TIME----
Note: The first inference is slow because it includes loading the model into Edge TPU memory.
578.49 ms
477.05 ms
498.25 ms
493.09 ms
551.80 ms
-------RESULTS--------
vase
id: 85
score: 0.6933819055557251
bbox: BBox(xmin=41, ymin=35, xmax=521, ymax=591)
The image is the same using for SSD MobileNet v2 example for object detection.
The ssdlite MobileDet was downloaded from the following site:
While using detection example with ssd mobilenet v2 everything works fine, when i load ssdlite mobiledet and using int8 quantized model, i get the following error:
On the other hand, when using fp32, the example code is running but the detection result is false:
The image is the same using for SSD MobileNet v2 example for object detection.
The ssdlite MobileDet was downloaded from the following site: