TypeError: Interpreter._get_tensor_details() missing 1 required positional argument: 'subgraph_index'

[Sukeysun]

Issue Type

Bug

OS

Ubuntu

OS architecture

x86_64

Programming Language

Python

Framework

ONNX

Download URL for tflite file

https://storage.googleapis.com/mediapipe-models/pose_landmarker/pose_landmarker_full/float16/latest/pose_landmarker_full.task unzip pose_landmarker_full.task

Convert Script

 tflite2tensorflow \
  --model_path ./pose_detector.tflite \
  --flatc_path ../flatbuffers/build/flatc \
  --schema_path ../schema.fbs \
  --model_output_path pose_detection \
  --output_pb

Description

I want to convert the pose_detector.tflite file to onnx file. I follow the step in readme: Step 1 : Generating saved_model and FreezeGraph (.pb)

 tflite2tensorflow \
  --model_path ./pose_detector.tflite \
  --flatc_path ../flatbuffers/build/flatc \
  --schema_path ../schema.fbs \
  --model_output_path pose_detection \
  --output_pb

Relevant Log Output

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ op: Placeholder
{'dtype': <class 'numpy.float32'>,
 'index': 0,
 'name': 'input_1',
 'quantization': (0.0, 0),
 'quantization_parameters': {'quantized_dimension': 0,
                             'scales': array([], dtype=float32),
                             'zero_points': array([], dtype=int32)},
 'shape': array([  1, 224, 224,   3], dtype=int32),
 'shape_signature': array([  1, 224, 224,   3], dtype=int32),
 'sparsity_parameters': {}}
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ op: PAD
{'builtin_options': {},
 'builtin_options_type': 'PadOptions',
 'custom_options_format': 'FLEXBUFFERS',
 'inputs': [0, 1],
 'opcode_index': 0,
 'outputs': [2]}
------------ 1
Traceback (most recent call last):
  File "/home/ai-server/.local/bin/tflite2tensorflow", line 819, in make_graph
    paddings_array = tensors[op['inputs'][1]]
                     ~~~~~~~^^^^^^^^^^^^^^^^^
KeyError: 1

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/ai-server/.local/bin/tflite2tensorflow", line 6615, in <module>
    main()
  File "/home/ai-server/.local/bin/tflite2tensorflow", line 5882, in main
    TFLite_Detection_PostProcess_flg = make_graph(
                                       ^^^^^^^^^^^
  File "/home/ai-server/.local/bin/tflite2tensorflow", line 822, in make_graph
    paddings_detail = interpreter._get_tensor_details(op['inputs'][1])
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
TypeError: Interpreter._get_tensor_details() missing 1 required positional argument: 'subgraph_index'

### Source code for simple inference testing code

_No response_

[PINTO0309]

docker run -it --rm \
-v `pwd`:/home/user/workdir \
ghcr.io/pinto0309/tflite2tensorflow:latest

tflite2tensorflow \
--model_path ./pose_detector.tflite \
--flatc_path ../flatc \
--schema_path ../schema.fbs \
--output_pb

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ op: RESHAPE
{'builtin_options_type': 'NONE',
 'custom_options_format': 'FLEXBUFFERS',
 'inputs': [439, 392],
 'opcode_index': 8,
 'outputs': [440]}
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ op: CONCATENATION
{'builtin_options': {'axis': 1, 'fused_activation_function': 'NONE'},
 'builtin_options_type': 'ConcatenationOptions',
 'custom_options_format': 'FLEXBUFFERS',
 'inputs': [393, 416, 440],
 'opcode_index': 9,
 'outputs': [441]}
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
outputs:
{'dtype': <class 'numpy.float32'>,
 'index': 441,
 'name': 'Identity',
 'quantization': (0.0, 0),
 'quantization_parameters': {'quantized_dimension': 0,
                             'scales': array([], dtype=float32),
                             'zero_points': array([], dtype=int32)},
 'shape': array([   1, 2254,   12], dtype=int32),
 'shape_signature': array([   1, 2254,   12], dtype=int32),
 'sparsity_parameters': {}}
{'dtype': <class 'numpy.float32'>,
 'index': 429,
 'name': 'Identity_1',
 'quantization': (0.0, 0),
 'quantization_parameters': {'quantized_dimension': 0,
                             'scales': array([], dtype=float32),
                             'zero_points': array([], dtype=int32)},
 'shape': array([   1, 2254,    1], dtype=int32),
 'shape_signature': array([   1, 2254,    1], dtype=int32),
 'sparsity_parameters': {}}
TensorFlow/Keras model building process complete!
saved_model / .pb output started ====================================================
saved_model / .pb output complete!

saved_model_cli show --dir saved_model/ --all

MetaGraphDef with tag-set: 'serve' contains the following SignatureDefs:

signature_def['serving_default']:
  The given SavedModel SignatureDef contains the following input(s):
    inputs['input_1'] tensor_info:
        dtype: DT_FLOAT
        shape: (1, 224, 224, 3)
        name: input_1:0
  The given SavedModel SignatureDef contains the following output(s):
    outputs['Identity'] tensor_info:
        dtype: DT_FLOAT
        shape: (1, 2254, 12)
        name: Identity:0
    outputs['Identity_1'] tensor_info:
        dtype: DT_FLOAT
        shape: (1, 2254, 1)
        name: Identity_1:0
  Method name is: tensorflow/serving/predict

[zhenhao-huang]

It's hard to pull down the Docker container.

[PINTO0309]

python -m tf2onnx.convert \
--opset 11 \
--tflite pose_landmarks_detector.tflite \
--output pose_landmarks_detector.onnx \
--inputs-as-nchw input_1 \
--dequantize

onnxsim pose_landmarks_detector.onnx pose_landmarks_detector.onnx