Andy Hsieh
Andy Hsieh
About
- Username
- Andy Hsieh
- Joined
- Visits
- 1,170
- Last Active
- Roles
- Member, ModeratorsKL520, ModeratorsKL720, ModeratorsAIModel
Comments
-
@Evelyn Hi Evelyn, 看您的錯誤訊息是在 kears model 轉成 .onnx 時就出錯了。 在 Keras 上我們的轉換器支援來自 Keras 2.2.4 的 model，請確認您的版本。 目前在hard ware上 Conv 的支援不包含 Conv1D，至於 padding='casual' 為 Kears 中 conv1D 的參數所以相對來說…
-
@Praveen Hi Praveen, Yes, the model you provided ('yolov5s-noupsample_convert.zip') can be used in KL720.🙂
-
@Praveen Hi Praveen, Your question has two parts: First, the model you are using has unsupported nodes. Second, if you want to use your model, you can only use KL720 to make inferences. KL520 currently does not support Sigmoid operator. Before you…
-
@oded Hi Oded, You can try putting your optimized model ONNX file under /data1/ (/data1/model.onnx) in docker to successfully use ktc.kneron_inference and complete E2E simulation.
-
@oded Hi Oded, I'd like to confirm your question when you said, "we get the segmentation error when running inference on the ONNX model and the bie model. The NEF model doesn't cause a crash." Do you mean that you have successfully conver…
-
@oded Hi oded, I don't know if you have optimized ONNX with the onnx2onnx API after you transferred the ONNX. After the optimization is complete, you must perform quantification. If you are worried about accuracy, you can use our E2E simulator for…
-
@Jasmine Hi Jasmine, 您的問題有三個部份： 1.Tool chain docker，您可以使用Kneron的tool chain docker將您的Model從 .onnx 轉成 .nef，這個步驟可確認您的Model架構是否支援Kneron硬體。 2.因為您使用的host_lib為Kneron的舊架構沒有特別更新維護,建議您可以直…
-
@Fred Hi Fred, 您給的 "yolov3-tiny.onnx" model output shape 不正確，您可以修改 model output shape 修改方式如下，修改後您就可以繼續完成指令 python /workspace/libs/ONNX_Convertor/optimizer_scripts/onnx1_4to1_6.py yolov3-tiny_modified_out…
-
Hi Richard, 在一般狀況下，直接建議您使用quantize_mode="default". 特殊情況下如果你的Model最後有sigmoid operator我們會將Model中有sigmoid operator部份先切割掉，再post process的時候處理。analysis function中參數quantize_mode才會建議您使用quant…
-
@Jasmine Hi Jasmine, 我們使用您提供tftargetcut_v2_opt.onnx的input_params.json執行以下指令是可以正確產生nef檔。 這邊幫您回答您的問題 我之前有跑過耐能給的範例（LittleNet）是可以正確執行指令產生nef檔的。（但執行compilerIpevaluator_720.sh的時候一樣會…