Jiyuan Liu
Jiyuan Liu
About
- Username
- Jiyuan Liu
- Joined
- Visits
- 22
- Last Active
- Roles
- Member
Comments
-
Hi Tim, fpAnalyserBatchCompiler.sh.x shell script is no longer available since v0.10.0. And I don't think you can find it in the current manual. You can find this information on the major changes list in section 0. The config you copied seems too o…
-
Hi Ellen, 上面提到的 forum 中的 yolo example 已经被重新编辑整理成了 step by step 的教程，见 http://doc.kneron.com/docs/#toolchain/yolo_example/ 根据你的链接生成的 yolov3-tiny 模型完全没有问题，不如说应该和我们那个 step by step 的教程提到的 repo 是一…
-
I think the issue here is that your pth only has the weight without the network structrure. Please double check how you get the pth file. If possible, try using `torch.onnx` to export the onnx yourself instead of using this pth. Using pth file is n…
-
Hi Ricardo, If you are converting yolov5s, I'm sorry that those slice operators cannot be converted correctly for now. We plan to switch to opset 11 since toolchain version 0.14, which should come out around May. If you are curious about how to sup…
-
Hi Ricardo, I'm sorry currently our toolchain only supports opset 9 ONNX. We're upgrading our toolchain to support opset 11. But it would be a later version. For now, we have a workaround. You can add the flag --align-corner while using pytorch2onn…
-
https://www.kneron.com/forum/discussion/comment/102#Comment_102 Okay, I'll check it.
-
https://www.kneron.com/forum/discussion/comment/98#Comment_98 I think Max's issue is solved in this thread: https://www.kneron.com/forum/discussion/41/runtimeerror-when-converting-onnx-model-using-pytorch2onnx-py#latest This was caused by a bug in t…
-
Hi Max, The bug should be fixed now. You can run `git pull` under /workspace/libs/ONNX_Converter to update the converter. Please have a try. Thank you for your patience.
-
https://www.kneron.com/forum/discussion/comment/66#Comment_66 Thank you for the model. I'm still checking it. It's a little bit busy recently. Sorry for the delay.
-
Hi, Please try pulling kneron/toolchain:520 for the latest 520 toolchain. By the way, please be careful using the mirror sites. Mirror sites are not always fully synced with the official docker hub site.
-
https://www.kneron.com/forum/discussion/comment/60#Comment_60 Hi Max, Would you share the model before pytorch2onnx.py? I could help you locate and solve the issue. The model you shared is after onnx2onnx which already has the dummy BatchNormalizati…
-
https://www.kneron.com/forum/discussion/comment/56#Comment_56 Hi Max, Yes, it should have set the shape. Then, it might be something wrong with the export function or the converter. Could you please share the model you exported? Dummy weight is oka…
-
Hi Max, It seems the Sigmoid node called 'output' in your model has the wrong output shape in the second dimension. In your model -9223372036854775808 is given but it should be 2.