search

google-coral / tflite

Examples using TensorFlow Lite API to run inference on Coral devices
https://coral.withgoogle.com
Apache License 2.0
181 stars 67 forks source link

RuntimeError: Internal: Unsupported data type: 0Node number 1 (EdgeTpuDelegateForCustomOp) failed to prepare. #9

Closed lsq314 closed 4 years ago

lsq314 commented 4 years ago

Hi all,

I am using the Coral Dev Board with the Mendel OS 4.0 (Day) version and tflite_runtime version 2.1.0. However, when I run the following example (officially provided) python3 classify_image.py --model models/mobilenet_v2_1.0_224_inat_bird_quant_edgetpu.tflite --labels models/inat_bird_labels.txt --input images/parrot.jpg I get the following error message. Traceback (most recent call last): File "classify_image.py", line 122, in <module> main() File "classify_image.py", line 100, in main interpreter.allocate_tensors() File "/home/mendel/.local/lib/python3.7/site-packages/tflite_runtime/interpreter.py", line 242, in allocate_tensors return self._interpreter.AllocateTensors() File "/home/mendel/.local/lib/python3.7/site-packages/tflite_runtime/interpreter_wrapper.py", line 115, in AllocateTensors return _interpreter_wrapper.InterpreterWrapper_AllocateTensors(self) RuntimeError: Internal: Unsupported data type: 0Node number 1 (EdgeTpuDelegateForCustomOp) failed to prepare.

Meanwhile, when I run the detection's example. The error is similar. Could someone please figure this out?

lsq314 commented 4 years ago

Plus, I could run the inference using the "edgetpu api" well.

Namburger commented 4 years ago

@lsq314 have you upgraded your libedgetpu runtime to version 13?

Namburger commented 4 years ago

Ahh, can you also upgrade libedgetpu1-std also?

lsq314 commented 4 years ago

@Namburger I find that no. I will try this.

lsq314 commented 4 years ago

@Namburger Thank you very much. It's been figured out.