You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
i compiled MNNConvert(2024.04.17 pulled from master branch)on ubuntu20.04. it convert onnx model of image encode part of SAM successfully. but when i tried to convert decode part, it failed.
(both onnx models work well under onnxruntime python.)
convert command ./MNNConvert -f ONNX --modelFile the_rest.onnx --MNNModel the_rest.mnn --bizCode MNN
failed info:
The device support i8sdot:0, support fp16:0, support i8mm: 0
Start to Convert Other Model Format To MNN Model..., target version: 2.8
[16:37:20] /home/amin_u_20/work/MNN-master/tools/converter/source/onnx/onnxConverter.cpp:46: ONNX Model ir version: 8
[16:37:20] /home/amin_u_20/work/MNN-master/tools/converter/source/onnx/onnxConverter.cpp:47: ONNX Model opset version: 17
Start to Optimize the MNN Net...
Segmentation fault
i compiled MNNConvert(2024.04.17 pulled from master branch)on ubuntu20.04. it convert onnx model of image encode part of SAM successfully. but when i tried to convert decode part, it failed.
(both onnx models work well under onnxruntime python.)
convert command
./MNNConvert -f ONNX --modelFile the_rest.onnx --MNNModel the_rest.mnn --bizCode MNN
failed info:
model url: https://github.com/AminGeng/Models/blob/main/the_rest.onnx
The text was updated successfully, but these errors were encountered: