[TODO] Add 4 channels of image data to the sample data for quantization

[PINTO0309]

Issue Type

Others

OS

Linux

onnx2tf version number

1.14.3

onnx version number

1.13.1

onnxruntime version number

1.15.0

onnxsim (onnx_simplifier) version number

0.4.17

tensorflow version number

2.13.0

Download URL for ONNX

magic_touch.onnx.zip

Parameter Replacement JSON

N/A

Description

1. Personal
2. I want to automatically quantize 4 channels of image data.
3. procedure

   onnx2tf -i magic_touch.onnx -osd -coion -cotof -oiqt -qt per-tensor
   

   saved_model output started ==========================================================
   saved_model output complete!
   Float32 tflite output complete!
   Float16 tflite output complete!
   Input signature information for quantization
   signature_name: serving_default
   input_name.0: input shape: (1, 512, 512, 4) dtype: <dtype: 'float32'>
   Dynamic Range Quantization tflite output complete!
   ERROR: For models that have multiple input OPs and need to perform INT8 quantization calibration using non-rgb-image input tensors, specify the calibration data with --quant_calib_input_op_name_np_data_path. model_input[n].shape: (1, 512, 512, 4)