rknn-toolkit2 icon indicating copy to clipboard operation
rknn-toolkit2 copied to clipboard

Results 228 rknn-toolkit2 issues
Sort by recently updated
recently updated
newest added

Hello, I'm trying to inference YOLOv4 .rknn converted from .onnx with the help of rknn-toolkit2. It performs well on PC using test.py code from toolkit examples but fails on AIO-3568J...

板端环境rk3568,模型为yolov5,合并3个输出头有使用pow层,未量化,精度分析工具显示差异很大,模型及错误日志见[百度网盘,提取码:wr4d](https://pan.baidu.com/s/1N1pgPPWFwP9glhRGRllm4g) ![版本驱动信息](https://user-images.githubusercontent.com/24456969/196838132-578047cf-065f-49fc-bf49-3907f8bff4f0.png) ![精度错误位置](https://user-images.githubusercontent.com/24456969/196838147-d7a80162-7f98-41cc-8981-3847ddd49f24.png)

Hi. We are developing our own models with our invented neural operators. But I found only several old models can run on this NPU. I wonder if there is any...

I RKNN: [09:58:04.794] RKNN Runtime Information: librknnrt version: 1.3.0 (c193be371@2022-05-04T20:16:33) I RKNN: [09:58:04.794] RKNN Driver Information: version: 0.6.4 I RKNN: [09:58:04.794] RKNN Model Information: version: 1, toolkit version: 1.3.0-11912b58(compiler version:...

![image](https://user-images.githubusercontent.com/47803360/196376137-a423963d-e6c4-429d-a3b6-507e077cf699.png)

Hi, I'm trying to convert yolov3 onnx model to rknn model, but got fail. the model source is from the link here: https://docs.openvino.ai/latest/omz_models_model_yolo_v3_onnx.html#doxid-omz-models-model-yolo-v3-onnx and I cut off the postprocess and...

你好。 模型:osnet 环境:python=3.6,rknn-toolkit2=1.4 在将onnx文件转为rknn的test中,load_onnx时报如下错,我没有找到相关问题,所以这是什么原因导致的呢,希望能得到解答。 ![2022-10-12 17-59-00屏幕截图](https://user-images.githubusercontent.com/57828055/195316200-0b25b995-7f67-4a9a-95e8-375afa69ae1e.png)

--> Init runtime environment I target set by user is: RK3588 I Starting ntp or adb, target is RK3588 I Device [af3985ee17a6830c] not found in ntb device list. I Start...

您好,我看op支持文档上,onnx支持lstm转换,实际在转带lstm的模型时,发现段错误,原始模型为降噪模型DTLN模型,模型结构及原始模型如下: ![image](https://user-images.githubusercontent.com/26894108/189840087-e948074d-701b-47a4-9c1e-dd3027e17e4f.png) ![06A7210C20783555A49DEBBD8B62DE7D](https://user-images.githubusercontent.com/26894108/189840101-c11b0831-bda6-4074-a4ee-6ff0afc6d60b.jpg) [model_p1.zip](https://github.com/rockchip-linux/rknn-toolkit2/files/9554606/model_p1.zip)