AssertionError: only Conv2D and Conv3D is supported; raise Exception("convert failed node:{}, op_type is {}".format( Exception: convert failed node:onnx__Transpose_597, op_type is Conv

[Xdream007]

问题描述

使用openmmlab中的mmsegmentation得到了deeplabV3Plus模型用于语义分割，并使用mmdeploy将模型转化为.onnx文件；接着使用X2Paddle将.onnx转化为paddle模型（x2paddle --framework=onnx --model=end2end.onnx --save_dir=pd_model）出现错误： AssertionError: only Conv2D and Conv3D is supported; raise Exception("convert failed node:{}, op_type is {}".format( Exception: convert failed node:onnx__Transpose_597, op_type is Conv

• 错误信息 C:\Users\Mr.dream007\Documents\anaconda3\envs\paddle_env\python.exe D:/learning_code/paddlepaddle/Paddle-Lite/X2paddle_Lite.py model ir_version: 6, op version: 11 Shape inferencing ... Shape inferenced. Now, onnx2paddle support convert onnx model opset_verison [7, 8, 9, 10, 11, 12, 13, 14, 15], opset_verison of your onnx model is 11. Total nodes: 112 Nodes converting ... Converting node 199 ... Traceback (most recent call last): File "C:\Users\Mr.dream007\Documents\anaconda3\envs\paddle_env\lib\site-packages\x2paddle\op_mapper\onnx2paddle\opset_legacy.py", line 110, in run_mapping res = func(*args, **kwargs) File "C:\Users\Mr.dream007\Documents\anaconda3\envs\paddle_env\lib\site-packages\x2paddle\op_mapper\onnx2paddle\opset_legacy.py", line 2236, in Conv assert 2 <= convnd <= 3, 'only Conv2D and Conv3D is supported' AssertionError: only Conv2D and Conv3D is supported

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "D:/learning_code/paddlepaddle/Paddle-Lite/X2paddle_Lite.py", line 3, in onnx2paddle(model_path="end2end.onnx", save_dir="xwd_paddle_lite", convert_to_lite=True, lite_valid_places="arm", lite_model_type="naive_buffer") File "C:\Users\Mr.dream007\Documents\anaconda3\envs\paddle_env\lib\site-packages\x2paddle\convert.py", line 304, in onnx2paddle mapper = ONNXOpMapper(model) File "C:\Users\Mr.dream007\Documents\anaconda3\envs\paddle_env\lib\site-packages\x2paddle\op_mapper\onnx2paddle\onnx_op_mapper.py", line 52, in init func(node) File "C:\Users\Mr.dream007\Documents\anaconda3\envs\paddle_env\lib\site-packages\x2paddle\op_mapper\onnx2paddle\opset_legacy.py", line 112, in run_mapping raise Exception("convert failed node:{}, op_type is {}".format( Exception: convert failed node:onnx__Transpose_597, op_type is Conv

• 错误截图

具体信息

• 转换模型后用处 使用 Paddle 框架/ PaddleInference 推理预测 使用 Paddle-Lite 做移动端推理

• 模型来源 https://github.com/open-mmlab/mmsegmentation/blob/main/configs/mobilenet_v2/mobilenet-v2-d8_deeplabv3plus_4xb4-160k_ade20k-512x512.py

• 应用场景 尝试部署在安卓手机端或Jetson ORIN Nano上用于语义分割

• 版本信息 PaddlePaddle => 2.5.1.post112 X2Paddle => 1.4.1 paddlelite => 2.13rc0 ONNX => 1.14.0 来源框架版本(PyTorch/TF/ONNX/Caffe) => PyTorch 1.12.1

• 您的联系方式(邮箱/微信/电话)

• w.d.xu@zju.edu.cn

-onnx文件百度网盘地址： 链接：https://pan.baidu.com/s/1w6t6gYMNbyxJP9GWJPsf5w?pwd=abbb 提取码：abbb