yolov8s.onnx转rknn报错Meet unsupported MaxPool attribute 'dilations'!

[gottastudy]

您好！使用https://github.com/airockchip/ultralytics_yolov8 仓库中新建export.py代码如下： from ultralytics import YOLO

model = YOLO("yolov8s.pt") # 加载YOLO模型 success = model.export(format="rknn", opset=12) 导出的onnx模型含有三个maxpool，且每个maxpool层含有dilations属性，导致转rknn时报如下错误： I rknn building ... I RKNN: [21:41:17.350] compress = 0, conv_arith_fuse = 1, global_fuse = 1, multi-core-model-mode = I RKNN: librknnc version: 1.4.0 (3b4520e4f@2022-09-05T20:52:35) D RKNN: [21:41:17.385] RKNN is invoked ... D RKNN: [21:41:18.040] >>>>>> start: N4rknn23RKNNProfileAnalysisPassE D RKNN: [21:41:18.040] <<<<<<<< end: N4rknn23RKNNProfileAnalysisPassE D RKNN: [21:41:18.040] >>>>>> start: OpEmit E RKNN: [21:41:18.045] Meet unsupported MaxPool attribute 'dilations'!

但是利用贵仓库提供的onnx模型，对比发现将maxpool的dilations属性去掉了。请问该操作是如何实现的？