---onnx模型转换错误！

[复制链接] · 发表于 2022-3-7 22:46:16

我用如下网站下的代码，做了pretrain和提取了onnx模型文件
https://github.com/ultralytics/yolov5

然后放在了: /rknn-toolkit2-1.0.0_no_docker/examples/onnx/bztest
目录下做了转换处理：（bztest目录是我拷贝resnet50v2的副本改出来的）
（原本的resnet50v2/这个目录下跑python3做模型转换是没有问题的）

$ python3 test.py 跑出来如下错误结果，请问这是啥问题呢？

--> config model
done
--> Loading model
done
--> Building model
E build: Catch exception when building RKNN model!
E build: Traceback (most recent call last):
E build: File "rknn/api/rknn_base.py", line 2140, in rknn.api.rknn_base.RKNNBase.build
E build: File "rknn/api/graph_optimizer.py", line 2297, in rknn.api.graph_optimizer.GraphOptimizer.run
E build: File "rknn/api/graph_optimizer.py", line 2272, in rknn.api.graph_optimizer.GraphOptimizer._sparse_weight
E build: File "rknn/api/sparse_weight.py", line 521, in rknn.api.sparse_weight.sparse_weight
E build: File "rknn/api/sparse_weight.py", line 444, in rknn.api.sparse_weight._range_add
E build: ValueError: operands could not be broadcast together with shapes (80,) (3,)
Build resnet50 failed!

只看该作者 · 发表于 2022-3-8 10:09:57

太安静了不是