|
我使用的rk3568 ,对yolov5 模型进行剪枝微调后,onnx 文件不能转成rknn文件了。是什么原因呢?
*.pt文件是没有问题的。使用Netron 看onnx模型和剪枝前没有结构变化。
--> Loading model
E load_onnx: Catch exception when loading onnx model: /home/rknn-toolkit2_1.3/examples/onnx/yolov5/yolov5s.onnx!
E load_onnx: Traceback (most recent call last):
E load_onnx: File "rknn/api/rknn_base.py", line 1127, in rknn.api.rknn_base.RKNNBase.load_onnx
E load_onnx: File "/usr/local/lib/python3.8/dist-packages/onnx/__init__.py", line 115, in load_model
E load_onnx: model = load_model_from_string(s, format=format)
E load_onnx: File "/usr/local/lib/python3.8/dist-packages/onnx/__init__.py", line 152, in load_model_from_string
E load_onnx: return _deserialize(s, ModelProto())
E load_onnx: File "/usr/local/lib/python3.8/dist-packages/onnx/__init__.py", line 95, in _deserialize
E load_onnx: decoded = cast(Optional[int], proto.ParseFromString(s))
E load_onnx: google.protobuf.message.DecodeError: Error parsing message
Load model failed!
|
本帖子中包含更多资源
您需要 登录 才可以下载或查看,没有帐号?立即注册
x
|