rpi-object-tracking: Coral USB is not working
- Raspberry Pi Deep PanTilt version: 4 with 4GB & 64SD
- Python version: 3.7
- Operating System: Debian Buster
Description
Ran “rpi-deep-pantilt detect --edge-tpu --loglevel=INFO” after installing Coral USB per instruction & Googl’s (Note it is now 2.1 version Runtime)"
Error: “RuntimeError: Internal: Unsupported data type in custom op handler: 0Node number 2 (EdgeTpuDelegateForCustomOp) failed to prepare.”
It works without “–edge-tpu”
What I Did: multiple re-install & reboot & same result
Paste the command(s) you ran and the output.: "rpi-deep-pantilt detect --edge-tpu" and "rpi-deep-pantilt detect --edge-tpu --loglevel=INFO" ----generate same error.
Whole output of error from CMD
(.venv) pi@raspberrypi:~/rpi-deep-pantilt $ rpi-deep-pantilt detect --edge-tpu
INFO: Initialized TensorFlow Lite runtime.
Traceback (most recent call last):
File "/home/pi/rpi-deep-pantilt/.venv/bin/rpi-deep-pantilt", line 8, in <module>
sys.exit(main())
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/rpi_deep_pantilt/cli.py", line 107, in main
cli()
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/click/core.py", line 764, in __call__
return self.main(*args, **kwargs)
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/click/core.py", line 717, in main
rv = self.invoke(ctx)
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/click/core.py", line 1137, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/click/core.py", line 956, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/click/core.py", line 555, in invoke
return callback(*args, **kwargs)
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/rpi_deep_pantilt/cli.py", line 52, in detect
model = SSDMobileNet_V3_Coco_EdgeTPU_Quant()
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/rpi_deep_pantilt/detect/ssd_mobilenet_v3_coco.py", line 56, in __init__
self.tflite_interpreter.allocate_tensors()
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/tensorflow_core/lite/python/interpreter.py", line 244, in allocate_tensors
return self._interpreter.AllocateTensors()
File "/home/pi/rpi-deep-pantilt/.venv/lib/python3.7/site-packages/tensorflow_core/lite/python/interpreter_wrapper/tensorflow_wrap_interpreter_wrapper.py", line 106, in AllocateTensors
return _tensorflow_wrap_interpreter_wrapper.InterpreterWrapper_AllocateTensors(self)
RuntimeError: Internal: Unsupported data type in custom op handler: 0Node number 2 (EdgeTpuDelegateForCustomOp) failed to prepare.
If there was a crash, please include the traceback here.
About this issue
- Original URL
- State: closed
- Created 4 years ago
- Comments: 20 (7 by maintainers)
Complementing to IanShow15’s contribution, the following worked for my RPI4:
in your venv, install this
pip3 install https://dl.google.com/coral/python/tflite_runtime-2.1.0.post1-cp37-cp37m-linux_armv7l.whlthe same procedure as lanShow15, but at the last step replace line 50-55 with the following instead:
self.tflite_interpreter = tflite.Interpreter( model_path=self.model_path, experimental_delegates=[ tflite.load_delegate(self.EDGETPU_SHARED_LIB) ])Nam Vu, Great & appreciate your comment and link!!!
@Namburger
No worries / no apology necessary! Thank you for jumping in and helping everyone in this issue. ❤️
Let me know if there’s anything I can do to assist you in the meantime!
The Google GDE program operates behind Google NDAs (we don’t work for or otherwise represent Google though). GDEs often organize/participate in early access programs if you’re looking for extra feedback, testing, support before open sourcing. I have a couple USB accelerators, the Dev Board, and I’m more than happy to do bonkers things like try the SOM on an RPI running an aarch64 distribution like Mendel, Fedora, Arch, etc.
Shoot me an email hi@leighjohnson.me if you want to connect and chat more about working with the GDE program.
for anyone that needs a temporary fix, you can do the following:
pip3 install https://dl.google.com/coral/python/tflite_runtime-2.1.0-cp37-cp37m-linux_aarch64.whladd
import tflite_runtime.interpreter as tfliteat the topreplace line 50-55
self.tflite_interpreter = tf.lite.Interpreter( model_path=self.model_path, experimental_delegates=[ tf.lite.experimental.load_delegate(self.EDGETPU_SHARED_LIB) ] )with the following
self.tflite_interpreter = tflite.Interpreter( model_path=model_file, experimental_delegates=[ tflite.load_delegate(EDGETPU_SHARED_LIB) ])Thank you @namburger! 🙏 Appreciate the example code. I’ll fix this in my next release.
@leigh-johnson I mentioned you to our FAE team and our Developer Advocates, they’ll contact you if we need your helps!
@leigh-johnson
This is indeed non standard and we apologize. The issue is that libedgetpu started depending on tensorflow as a dependency, so you’d need the exact tensorflow commit that we used to build libedgetpu in order to be compatible. That’s why we packaged
tflite_runtimepackage that is built from the same commit aslibedgetpu.so. The deeper reason why this is so is because our library isn’t open source which means users cannot build their own.soto fit with the tensorflow package. We are working diligently on this issue. Thanks for this repoHey y’all, this should be fixed in versions >= 1.2.0 - let me know experience any issues!
I’m also looking into why
edgetpulibneeds to package its ownInterpreterandDelegatebase classes, this is non-standard.My understanding is that these interfaces are provided by TensorFlow’s
tensorflow.litelib, and thattf.lite.experimental.load_delegatecan be used to load a library that implements a custom delegate class. The custom delegate is responsible for registering a kernel node, which parses a TensorFlow graph and “claims” operations it knows how to execute.The
Unsupported data type in custom op handlererror raised by usinglibedgetpu’s shared object withtensorflow.litesmells like a SWIG typemap issue, but I’d need more context from Coral folks to get to the bottom of that issue.@Martin2kid The problem here is that this repo is using
tf.lite.experimental.load_delegatewhich is a tensorflow api:https://github.com/leigh-johnson/rpi-deep-pantilt/blob/master/rpi_deep_pantilt/detect/facessd_mobilenet_v2.py#L53I suggest using the
tflite_runtimeapi instead which is the reason why it was upgraded recently in the first place. You can see here as an example to do this: https://github.com/google-coral/tflite/blob/master/python/examples/detection/detect_image.py#L57