Hi experts, I’m trying to convert paddle ocr rec model to hef file. When I parse the model with below script
runner = ClientRunner(hw_arch=chosen_hw_arch)
hn, npz = runner.translate_onnx_model(
onnx_path,
onnx_model_name,
start_node_names=["x"],
#end_node_names=["p2o.Sigmoid.0"],
net_input_shapes={"x": [1, 3, 768, 768]},
)
I got below error message:
[info] Translation started on ONNX model ch_PP-OCRv3_rec
[info] Restored ONNX model ch_PP-OCRv3_rec (completion time: 00:00:00.09)
[info] Extracted ONNXRuntime meta-data for Hailo model (completion time: 00:00:01.62)
[warning] ONNX shape inference failed: [ONNXRuntimeError] : 1 : FAIL : Node (Concat_1) Op (Concat) [ShapeInferenceError] Can't merge shape info. Both inferred and declared dimension have values but they differ. Inferred=1 Declared=24 Dimension=2
[info] Simplified ONNX model for a parsing retry attempt (completion time: 00:00:06.39)
[warning] ONNX shape inference failed: [ONNXRuntimeError] : 1 : FAIL : Node (Concat_1) Op (Concat) [ShapeInferenceError] Can't merge shape info. Both inferred and declared dimension have values but they differ. Inferred=1 Declared=24 Dimension=2
Can anyone give me a input to fix this issue?