[onnx] Unimplemented operators for LeYOLO
Linked to issue : https://gitlab.eclipse.org/eclipse/aidge/aidge_onnx/-/issues/10
What commit version of aidge do you use
Package Version
aidge_backend_cpu 0.3.2.dev17+g2a745b7 aidge_backend_opencv 0.0.4.dev34+g45728a6 aidge_core 0.2.3.dev544+g07c12bbb aidge_export_cpp 0.1.3.dev68+gb1f0cbe aidge_onnx 0.3.1.dev13+g448a776 colorama 0.4.6 Jinja2 3.1.4 MarkupSafe 2.1.5 numpy 1.24.4 onnx 1.17.0 pip 24.3.1 protobuf 5.28.3 setuptools 74.1.2 wheel 0.44.0
Problem description
Operators Resize, Reshape with parameter allowzero, Split, Shape are imported as GenericOperator, i.e. as unimplemented operators.
Is this a regression ? No
Please provide logs in the form of a code block
(env_aidge) farges@WDTIS181H:/mnt/d/farges/Documents/LeYOLO/weights$ python3 script.py
Available backends :
{'export_serialize', 'cpu'}
- model_0_c_Conv (Conv)
- model_0_act_Sigmoid (Sigmoid)
- model_0_act_Mul (Mul)
- model_1_c_Conv (Conv)
- model_1_act_Sigmoid (Sigmoid)
- model_1_act_Mul (Mul)
- model_2_layers_layers_0_c_Conv (Conv)
- model_2_layers_layers_0_act_Sigmoid (Sigmoid)
- model_2_layers_layers_0_act_Mul (Mul)
- model_2_layers_layers_1_Conv (Conv)
- model_3_layers_layers_0_c_Conv (Conv)
- model_3_layers_layers_0_act_Sigmoid (Sigmoid)
- model_3_layers_layers_0_act_Mul (Mul)
- model_3_layers_layers_1_c_Conv (Conv)
- model_3_layers_layers_1_act_Sigmoid (Sigmoid)
- model_3_layers_layers_1_act_Mul (Mul)
- model_3_layers_layers_2_Conv (Conv)
- model_4_layers_layers_0_c_Conv (Conv)
- model_4_layers_layers_0_act_Sigmoid (Sigmoid)
- model_4_layers_layers_0_act_Mul (Mul)
- model_4_layers_layers_1_c_Conv (Conv)
- model_4_layers_layers_1_act_Sigmoid (Sigmoid)
- model_4_layers_layers_1_act_Mul (Mul)
- model_4_layers_layers_2_Conv (Conv)
- model_4_Add (Add)
- model_5_layers_layers_0_c_Conv (Conv)
- model_5_layers_layers_0_act_Sigmoid (Sigmoid)
- model_5_layers_layers_0_act_Mul (Mul)
- model_5_layers_layers_1_c_Conv (Conv)
- model_5_layers_layers_1_act_Sigmoid (Sigmoid)
- model_5_layers_layers_1_act_Mul (Mul)
- model_5_layers_layers_2_Conv (Conv)
- model_6_layers_layers_0_c_Conv (Conv)
- model_6_layers_layers_0_act_Sigmoid (Sigmoid)
- model_6_layers_layers_0_act_Mul (Mul)
- model_6_layers_layers_1_c_Conv (Conv)
- model_6_layers_layers_1_act_Sigmoid (Sigmoid)
- model_6_layers_layers_1_act_Mul (Mul)
- model_6_layers_layers_2_Conv (Conv)
- model_6_Add (Add)
- model_7_layers_layers_0_c_Conv (Conv)
- model_7_layers_layers_0_act_Sigmoid (Sigmoid)
- model_7_layers_layers_0_act_Mul (Mul)
- model_7_layers_layers_1_c_Conv (Conv)
- model_7_layers_layers_1_act_Sigmoid (Sigmoid)
- model_7_layers_layers_1_act_Mul (Mul)
- model_7_layers_layers_2_Conv (Conv)
- model_7_Add (Add)
- model_8_layers_layers_0_c_Conv (Conv)
- model_8_layers_layers_0_act_Sigmoid (Sigmoid)
- model_8_layers_layers_0_act_Mul (Mul)
- model_8_layers_layers_1_c_Conv (Conv)
- model_8_layers_layers_1_act_Sigmoid (Sigmoid)
- model_8_layers_layers_1_act_Mul (Mul)
- model_8_layers_layers_2_Conv (Conv)
- model_8_Add (Add)
- model_9_layers_layers_0_c_Conv (Conv)
- model_9_layers_layers_0_act_Sigmoid (Sigmoid)
- model_9_layers_layers_0_act_Mul (Mul)
- model_9_layers_layers_1_c_Conv (Conv)
- model_9_layers_layers_1_act_Sigmoid (Sigmoid)
- model_9_layers_layers_1_act_Mul (Mul)
- model_9_layers_layers_2_Conv (Conv)
- model_9_Add (Add)
- model_10_layers_layers_0_c_Conv (Conv)
- model_10_layers_layers_0_act_Sigmoid (Sigmoid)
- model_10_layers_layers_0_act_Mul (Mul)
- model_10_layers_layers_1_c_Conv (Conv)
- model_10_layers_layers_1_act_Sigmoid (Sigmoid)
- model_10_layers_layers_1_act_Mul (Mul)
- model_10_layers_layers_2_Conv (Conv)
- model_11_layers_layers_0_c_Conv (Conv)
- model_11_layers_layers_0_act_Sigmoid (Sigmoid)
- model_11_layers_layers_0_act_Mul (Mul)
- model_11_layers_layers_1_c_Conv (Conv)
- model_11_layers_layers_1_act_Sigmoid (Sigmoid)
- model_11_layers_layers_1_act_Mul (Mul)
- model_11_layers_layers_2_Conv (Conv)
- model_11_Add (Add)
- model_12_layers_layers_0_c_Conv (Conv)
- model_12_layers_layers_0_act_Sigmoid (Sigmoid)
- model_12_layers_layers_0_act_Mul (Mul)
- model_12_layers_layers_1_c_Conv (Conv)
- model_12_layers_layers_1_act_Sigmoid (Sigmoid)
- model_12_layers_layers_1_act_Mul (Mul)
- model_12_layers_layers_2_Conv (Conv)
- model_12_Add (Add)
- model_13_layers_layers_0_c_Conv (Conv)
- model_13_layers_layers_0_act_Sigmoid (Sigmoid)
- model_13_layers_layers_0_act_Mul (Mul)
- model_13_layers_layers_1_c_Conv (Conv)
- model_13_layers_layers_1_act_Sigmoid (Sigmoid)
- model_13_layers_layers_1_act_Mul (Mul)
- model_13_layers_layers_2_Conv (Conv)
- model_13_Add (Add)
- model_14_cv1_conv_Conv (Conv)
- model_14_cv1_act_Sigmoid (Sigmoid)
- model_14_cv1_act_Mul (Mul)
- model_14_m_MaxPool (MaxPool)
- model_14_m_1_MaxPool (MaxPool)
- model_14_m_2_MaxPool (MaxPool)
- model_14_Concat (Concat)
- model_14_cv2_conv_Conv (Conv)
- model_14_cv2_act_Sigmoid (Sigmoid)
- model_14_cv2_act_Mul (Mul)
- model_15_Constant (Constant)
val type: float32
- model_15_Resize (Resize | GenericOperator)
- coordinate_transformation_mode : b'asymmetric'
- cubic_coeff_a : -0.75
- mode : b'nearest'
- nearest_mode : b'floor'
- model_16_Concat (Concat)
- model_17_layers_layers_0_c_Conv (Conv)
- model_17_layers_layers_0_act_Sigmoid (Sigmoid)
- model_17_layers_layers_0_act_Mul (Mul)
- model_17_layers_layers_1_c_Conv (Conv)
- model_17_layers_layers_1_act_Sigmoid (Sigmoid)
- model_17_layers_layers_1_act_Mul (Mul)
- model_17_layers_layers_2_Conv (Conv)
- model_18_layers_layers_0_c_Conv (Conv)
- model_18_layers_layers_0_act_Sigmoid (Sigmoid)
- model_18_layers_layers_0_act_Mul (Mul)
- model_18_layers_layers_1_c_Conv (Conv)
- model_18_layers_layers_1_act_Sigmoid (Sigmoid)
- model_18_layers_layers_1_act_Mul (Mul)
- model_18_layers_layers_2_Conv (Conv)
- model_18_Add (Add)
- model_19_layers_layers_0_c_Conv (Conv)
- model_19_layers_layers_0_act_Sigmoid (Sigmoid)
- model_19_layers_layers_0_act_Mul (Mul)
- model_19_layers_layers_1_c_Conv (Conv)
- model_19_layers_layers_1_act_Sigmoid (Sigmoid)
- model_19_layers_layers_1_act_Mul (Mul)
- model_19_layers_layers_2_Conv (Conv)
- model_19_Add (Add)
- model_20_Constant (Constant)
val type: float32
- model_20_Resize (Resize | GenericOperator)
- coordinate_transformation_mode : b'asymmetric'
- cubic_coeff_a : -0.75
- mode : b'nearest'
- nearest_mode : b'floor'
- model_21_Concat (Concat)
- model_22_layers_layers_0_c_Conv (Conv)
- model_22_layers_layers_0_act_Sigmoid (Sigmoid)
- model_22_layers_layers_0_act_Mul (Mul)
- model_22_layers_layers_1_Conv (Conv)
- model_23_layers_layers_0_c_Conv (Conv)
- model_23_layers_layers_0_act_Sigmoid (Sigmoid)
- model_23_layers_layers_0_act_Mul (Mul)
- model_23_layers_layers_1_c_Conv (Conv)
- model_23_layers_layers_1_act_Sigmoid (Sigmoid)
- model_23_layers_layers_1_act_Mul (Mul)
- model_23_layers_layers_2_Conv (Conv)
- model_23_Add (Add)
- model_24_layers_layers_0_c_Conv (Conv)
- model_24_layers_layers_0_act_Sigmoid (Sigmoid)
- model_24_layers_layers_0_act_Mul (Mul)
- model_24_layers_layers_1_c_Conv (Conv)
- model_24_layers_layers_1_act_Sigmoid (Sigmoid)
- model_24_layers_layers_1_act_Mul (Mul)
- model_24_layers_layers_2_Conv (Conv)
- model_24_Add (Add)
- model_25_c_Conv (Conv)
- model_25_act_Sigmoid (Sigmoid)
- model_25_act_Mul (Mul)
- model_26_Concat (Concat)
- model_27_layers_layers_0_c_Conv (Conv)
- model_27_layers_layers_0_act_Sigmoid (Sigmoid)
- model_27_layers_layers_0_act_Mul (Mul)
- model_27_layers_layers_1_c_Conv (Conv)
- model_27_layers_layers_1_act_Sigmoid (Sigmoid)
- model_27_layers_layers_1_act_Mul (Mul)
- model_27_layers_layers_2_Conv (Conv)
- model_28_layers_layers_0_c_Conv (Conv)
- model_28_layers_layers_0_act_Sigmoid (Sigmoid)
- model_28_layers_layers_0_act_Mul (Mul)
- model_28_layers_layers_1_c_Conv (Conv)
- model_28_layers_layers_1_act_Sigmoid (Sigmoid)
- model_28_layers_layers_1_act_Mul (Mul)
- model_28_layers_layers_2_Conv (Conv)
- model_28_Add (Add)
- model_29_layers_layers_0_c_Conv (Conv)
- model_29_layers_layers_0_act_Sigmoid (Sigmoid)
- model_29_layers_layers_0_act_Mul (Mul)
- model_29_layers_layers_1_c_Conv (Conv)
- model_29_layers_layers_1_act_Sigmoid (Sigmoid)
- model_29_layers_layers_1_act_Mul (Mul)
- model_29_layers_layers_2_Conv (Conv)
- model_29_Add (Add)
- model_30_c_Conv (Conv)
- model_30_act_Sigmoid (Sigmoid)
- model_30_act_Mul (Mul)
- model_31_Concat (Concat)
- model_32_layers_layers_0_c_Conv (Conv)
- model_32_layers_layers_0_act_Sigmoid (Sigmoid)
- model_32_layers_layers_0_act_Mul (Mul)
- model_32_layers_layers_1_c_Conv (Conv)
- model_32_layers_layers_1_act_Sigmoid (Sigmoid)
- model_32_layers_layers_1_act_Mul (Mul)
- model_32_layers_layers_2_Conv (Conv)
- model_33_layers_layers_0_c_Conv (Conv)
- model_33_layers_layers_0_act_Sigmoid (Sigmoid)
- model_33_layers_layers_0_act_Mul (Mul)
- model_33_layers_layers_1_c_Conv (Conv)
- model_33_layers_layers_1_act_Sigmoid (Sigmoid)
- model_33_layers_layers_1_act_Mul (Mul)
- model_33_layers_layers_2_Conv (Conv)
- model_33_Add (Add)
- model_34_layers_layers_0_c_Conv (Conv)
- model_34_layers_layers_0_act_Sigmoid (Sigmoid)
- model_34_layers_layers_0_act_Mul (Mul)
- model_34_layers_layers_1_c_Conv (Conv)
- model_34_layers_layers_1_act_Sigmoid (Sigmoid)
- model_34_layers_layers_1_act_Mul (Mul)
- model_34_layers_layers_2_Conv (Conv)
- model_34_Add (Add)
- model_35_cv2_0_cv2_0_0_conv_Conv (Conv)
- model_35_cv2_0_cv2_0_0_act_Sigmoid (Sigmoid)
- model_35_cv2_0_cv2_0_0_act_Mul (Mul)
- model_35_cv2_0_cv2_0_1_conv_Conv (Conv)
- model_35_cv2_0_cv2_0_1_act_Sigmoid (Sigmoid)
- model_35_cv2_0_cv2_0_1_act_Mul (Mul)
- model_35_cv2_0_cv2_0_2_conv_Conv (Conv)
- model_35_cv2_0_cv2_0_2_act_Sigmoid (Sigmoid)
- model_35_cv2_0_cv2_0_2_act_Mul (Mul)
- model_35_cv2_0_cv2_0_3_Conv (Conv)
- model_35_cv3_0_cv3_0_0_conv_Conv (Conv)
- model_35_cv3_0_cv3_0_0_act_Sigmoid (Sigmoid)
- model_35_cv3_0_cv3_0_0_act_Mul (Mul)
- model_35_cv3_0_cv3_0_1_conv_Conv (Conv)
- model_35_cv3_0_cv3_0_1_act_Sigmoid (Sigmoid)
- model_35_cv3_0_cv3_0_1_act_Mul (Mul)
- model_35_cv3_0_cv3_0_2_conv_Conv (Conv)
- model_35_cv3_0_cv3_0_2_act_Sigmoid (Sigmoid)
- model_35_cv3_0_cv3_0_2_act_Mul (Mul)
- model_35_cv3_0_cv3_0_3_Conv (Conv)
- model_35_Concat (Concat)
- model_35_cv2_1_cv2_1_0_conv_Conv (Conv)
- model_35_cv2_1_cv2_1_0_act_Sigmoid (Sigmoid)
- model_35_cv2_1_cv2_1_0_act_Mul (Mul)
- model_35_cv2_1_cv2_1_1_conv_Conv (Conv)
- model_35_cv2_1_cv2_1_1_act_Sigmoid (Sigmoid)
- model_35_cv2_1_cv2_1_1_act_Mul (Mul)
- model_35_cv2_1_cv2_1_2_conv_Conv (Conv)
- model_35_cv2_1_cv2_1_2_act_Sigmoid (Sigmoid)
- model_35_cv2_1_cv2_1_2_act_Mul (Mul)
- model_35_cv2_1_cv2_1_3_Conv (Conv)
- model_35_cv3_1_cv3_1_0_conv_Conv (Conv)
- model_35_cv3_1_cv3_1_0_act_Sigmoid (Sigmoid)
- model_35_cv3_1_cv3_1_0_act_Mul (Mul)
- model_35_cv3_1_cv3_1_1_conv_Conv (Conv)
- model_35_cv3_1_cv3_1_1_act_Sigmoid (Sigmoid)
- model_35_cv3_1_cv3_1_1_act_Mul (Mul)
- model_35_cv3_1_cv3_1_2_conv_Conv (Conv)
- model_35_cv3_1_cv3_1_2_act_Sigmoid (Sigmoid)
- model_35_cv3_1_cv3_1_2_act_Mul (Mul)
- model_35_cv3_1_cv3_1_3_Conv (Conv)
- model_35_Concat_1 (Concat)
- model_35_cv2_2_cv2_2_0_conv_Conv (Conv)
- model_35_cv2_2_cv2_2_0_act_Sigmoid (Sigmoid)
- model_35_cv2_2_cv2_2_0_act_Mul (Mul)
- model_35_cv2_2_cv2_2_1_conv_Conv (Conv)
- model_35_cv2_2_cv2_2_1_act_Sigmoid (Sigmoid)
- model_35_cv2_2_cv2_2_1_act_Mul (Mul)
- model_35_cv2_2_cv2_2_2_conv_Conv (Conv)
- model_35_cv2_2_cv2_2_2_act_Sigmoid (Sigmoid)
- model_35_cv2_2_cv2_2_2_act_Mul (Mul)
- model_35_cv2_2_cv2_2_3_Conv (Conv)
- model_35_cv3_2_cv3_2_0_conv_Conv (Conv)
- model_35_cv3_2_cv3_2_0_act_Sigmoid (Sigmoid)
- model_35_cv3_2_cv3_2_0_act_Mul (Mul)
- model_35_cv3_2_cv3_2_1_conv_Conv (Conv)
- model_35_cv3_2_cv3_2_1_act_Sigmoid (Sigmoid)
- model_35_cv3_2_cv3_2_1_act_Mul (Mul)
- model_35_cv3_2_cv3_2_2_conv_Conv (Conv)
- model_35_cv3_2_cv3_2_2_act_Sigmoid (Sigmoid)
- model_35_cv3_2_cv3_2_2_act_Mul (Mul)
- model_35_cv3_2_cv3_2_3_Conv (Conv)
- model_35_Concat_2 (Concat)
- model_35_Constant (Constant)
val type: int64
- model_35_Constant_1 (Constant)
val type: int64
- model_35_Constant_2 (Constant)
val type: int64
Warning: unsupported attribute(s): dict_keys(['allowzero']) for operator 'Reshape' with opset 17.
This node will be filled by a GenericOperator.
- model_35_Reshape (Reshape | GenericOperator)
- allowzero : 0
Warning: unsupported attribute(s): dict_keys(['allowzero']) for operator 'Reshape' with opset 17.
This node will be filled by a GenericOperator.
- model_35_Reshape_1 (Reshape | GenericOperator)
- allowzero : 0
Warning: unsupported attribute(s): dict_keys(['allowzero']) for operator 'Reshape' with opset 17.
This node will be filled by a GenericOperator.
- model_35_Reshape_2 (Reshape | GenericOperator)
- allowzero : 0
- model_35_Concat_3 (Concat)
- Constant_278 (Constant)
val type: int64
Warning: an error occured when trying to load node model_35_Split_output_0 of type split.
Loading node using a generic operator.
Please report this issue at https://gitlab.eclipse.org/eclipse/aidge/aidge_onnx
by providing your ONNX model and the following error:
ONNX_NODE_CONVERTER_ returned: 'split'
- model_35_Split (Split | GenericOperator)
- axis : 1
- model_35_dfl_Constant (Constant)
val type: int64
Warning: unsupported attribute(s): dict_keys(['allowzero']) for operator 'Reshape' with opset 17.
This node will be filled by a GenericOperator.
- model_35_dfl_Reshape (Reshape | GenericOperator)
- allowzero : 0
- model_35_dfl_Transpose (Transpose)
- model_35_dfl_Softmax (Softmax)
- model_35_dfl_conv_Conv (Conv)
- model_35_dfl_Constant_1 (Constant)
val type: int64
Warning: unsupported attribute(s): dict_keys(['allowzero']) for operator 'Reshape' with opset 17.
This node will be filled by a GenericOperator.
- model_35_dfl_Reshape_1 (Reshape | GenericOperator)
- allowzero : 0
Warning: an error occured when trying to load node model_35_Shape_output_0 of type shape.
Loading node using a generic operator.
Please report this issue at https://gitlab.eclipse.org/eclipse/aidge/aidge_onnx
by providing your ONNX model and the following error:
ONNX_NODE_CONVERTER_ returned: __init__(): incompatible constructor arguments. The following argument types are supported:
1. aidge_core.aidge_core.Node(op: Aidge::Operator, name: str = '')
Invoked with: Node(name='', optype='Shape', parents: [0], children: [[]]); kwargs: name='model_35_Shape_output_0'
- model_35_Shape (Shape | GenericOperator)
- model_35_Constant_3 (Constant)
val type: int64
- model_35_Gather (Gather)
- model_35_Constant_4 (Constant)
val type: int64
- model_35_Constant_5 (Constant)
val type: int64
- model_35_Add (Add)
- model_35_Constant_6 (Constant)
val type: int64
- model_35_Div (Div)
- model_35_Constant_7 (Constant)
val type: int64
- model_35_Mul (Mul)
- model_35_Slice (Slice)
- model_35_Constant_8 (Constant)
val type: int64
- model_35_Mul_1 (Mul)
- model_35_Slice_1 (Slice)
- model_35_Constant_9 (Constant)
val type: float32
- model_35_Sub (Sub)
- model_35_Constant_10 (Constant)
val type: float32
- model_35_Add_1 (Add)
- model_35_Add_2 (Add)
- model_35_Constant_11 (Constant)
val type: float32
- model_35_Div_1 (Div)
- model_35_Sub_1 (Sub)
- model_35_Concat_4 (Concat)
- model_35_Constant_12 (Constant)
val type: float32
- model_35_Mul_2 (Mul)
- model_35_Sigmoid (Sigmoid)
- model_35_Concat_5 (Concat)
(env_aidge) farges@WDTIS181H:/mnt/d/farges/Documents/LeYOLO/weights$Reproducible example code
You can attach code or use code block
import aidge_core
#import aidge_backend_cpu
import aidge_onnx
import numpy as np
#import matplotlib.pyplot as pl
print("Available backends :")
ab = aidge_core.Tensor.get_available_backends()
print(ab)
model = aidge_onnx.load_onnx("LeYOLOSmall.onnx")
model.save("myModel")Edited by Jean-Loup Farges