Problem with .onnx model loading
I am trying to run the custom model (yolov3) and I uploaded the weights according to the training example in jupyter notebook. Unfortunately, I have a strange problem with model creation, which does not allow me to proceed with the working flow.
The model weights are from github source:
https://github.com/onnx/models/tree/main/vision/object_detection_segmentation/yolov3
.
The error message is:
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
/home/orange/project/aidge/examples/tutorials/Aidge_tutorial/load_and_run.ipynb Cell 13 line 1
----> 1 model = aidge_onnx.load_onnx("/home/orange/project/aidge/examples/tutorials/Aidge_tutorial/yolov3-10.onnx")
2 #("/home/orange/project/aidge/examples/tutorials/Aidge_tutorial/yolov3-12.onnx")
File ~/project/aidge/.venv/lib/python3.10/site-packages/aidge_onnx/onnx_loader.py:31, in load_onnx(filename, verbose)
29 # Load the ONNX model
30 model = onnx.load(filename)
---> 31 return _load_onnx2graphview(model, verbose)
File ~/project/aidge/.venv/lib/python3.10/site-packages/aidge_onnx/onnx_loader.py:75, in _load_onnx2graphview(model, verbose)
73 model_inputs[node_name][id_learnable_parameter] = model_producers[input_node]
74 id_learnable_parameter+=1
---> 75 model_nodes[node_name] = ONNX_NODE_CONVERTER_[onnx_node.op_type.lower()](onnx_node, model_inputs[node_name])
77 # Add nodes to the model_inputs dict
78 for onnx_node in model.graph.node:
File ~/project/aidge/.venv/lib/python3.10/site-packages/aidge_onnx/node_converter/converters/conv.py:59, in import_conv(onnx_node, input_nodes)
57 in_channels = input_nodes[1].get_operator().output(0).dims()[1]
58 if f"Conv{len(kernel_dims)}D" in dir(aidge_core):
---> 59 conv_node = aidge_core.__getattribute__(f"Conv{len(kernel_dims)}D")(
60 in_channels,
61 out_channels,
62 kernel_dims,
...
TypeError: Conv2D(): incompatible function arguments. The following argument types are supported:
1. (in_channels: int, out_channels: int, kernel_dims: List[int], name: str = '', stride_dims: List[int] = [1, 1], dilation_dims: List[int] = [1, 1]) -> aidge_core.aidge_core.Node
Invoked with: 3, 32, [3, 3]; kwargs: name='convolution_output74', stride_dims=[1, 1], padding_dims=0, dilation_dims=[1, 1]