PaddlePaddle / PaddleDetection

Object Detection toolkit based on PaddlePaddle. It supports object detection, instance segmentation, multiple object tracking and real-time multi-person keypoint detection.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

模型需要转为onnx格式固定输入固定输出,但是PCB缺陷检测1个图可能不至1个缺陷,转了几次都不对

makaiyang opened this issue · comments

commented

问题确认 Search before asking

  • 我已经查询历史issue,没有类似需求。I have searched the issues and found no similar feature requests.

需求描述 Feature Description

模型需要转为onnx格式固定输入固定输出,但是PCB缺陷检测1个图可能不至1个缺陷,转了几次都不对
D:\ancanda\miini\envs\onnx\python.exe D:\ocr\Paddle2ONNX-develop\onnx_test.py
INFO:root:onnx model with info:8
INFO:root:onnx model with info:PaddlePaddle
INFO:root:onnx model with info:
INFO:root:input tensor name: image
INFO:root:input tensor data type: 1
INFO:root:input tensor with shape:[1, 3, 640, 640]
INFO:root:input tensor name: scale_factor
INFO:root:input tensor data type: 1
INFO:root:input tensor with shape:[1, 2]
INFO:root:output tensor name: multiclass_nms3_0.tmp_0
INFO:root:output tensor data type: 1
INFO:root:output tensor with shape:[-1, 6]
INFO:root:output tensor name: multiclass_nms3_0.tmp_2
INFO:root:output tensor data type: 6
INFO:root:output tensor with shape:[-1]
Process finished with exit code 0
使用的命令:(onnx) PS D:\ocr\Paddle2ONNX-develop> paddle2onnx --model_dir test --model_filename model.pdmodel --params_filename model.pdiparams --opset_version 11 --save_file PCB_detect.onnx --inp
ut_shape_dict "{'image':[1,3,640,640],'scale_factor':[1,2]}" --enable_onnx_checker True

是否愿意提交PR Are you willing to submit a PR?

  • Yes I'd like to help by submitting a PR!