There was insufficient consideration of the case where the MatMul is immediately after the 4D input layer

[PINTO0309]

Issue Type

Bug

OS

Ubuntu

OS architecture

x86_64

Programming Language

Python

Framework

OpenVINO, PyTorch, ONNX, TensorFlow, TensorFlowLite

Download URL for ONNX / OpenVINO IR

https://github.com/PINTO0309/openvino2tensorflow/files/9409671/model.zip

Convert Script

mo \
--framework onnx \
--input_model model.onnx \
--data_type FP32 \
--output_dir openvino/FP32 \
--model_name model

openvino2tensorflow \
--model_path openvino/FP32/model.xml \
--output_saved_model \
--output_pb \
--non_verbose \
--output_no_quant_float32_tflite

Description

Error occurs when MatMul is immediately after the 4D input layer. https://github.com/PINTO0309/openvino2tensorflow/pull/120#issuecomment-1224570231

Relevant Log Output

ValueError: Dimensions must be equal, but are 3 and 28 for '{{node tf.linalg.matmul/MatMul}} = BatchMatMulV2[T=DT_FLOAT, adj_x=false, adj_y=true](Placeholder, tf.linalg.matmul/MatMul/b)' with input shapes: [1,28,28,3], [10,28].
ERROR: Please refer to 6-7 in the README first. https://github.com/PINTO0309/openvino2tensorflow

Source code for simple inference testing code

mo \
--framework onnx \
--input_model model.onnx \
--data_type FP32 \
--output_dir openvino/FP32 \
--model_name model

openvino2tensorflow \
--model_path openvino/FP32/model.xml \
--output_saved_model \
--output_pb \
--non_verbose \
--output_no_quant_float32_tflite