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

Results 208 rknn-toolkit2 issues
Sort by recently updated
recently updated
newest added

![屏幕截图 2022-11-17 175150](https://user-images.githubusercontent.com/65838905/202415018-8e892a41-3f31-4166-8f83-a81999c6304a.png) 算法:YOLOV6s 模型文件(百度云盘):https://pan.baidu.com/s/1IatoQI5r3eC6jAsAjlnBdQ 提取码: ccgj

想把光流的模型,比如flowNet2弄到npu上运行,但correlation和warping算子不支持.

你好,请问[NanoPi R6S](https://wiki.friendlyelec.com/wiki/index.php/NanoPi_R6S/zh)这款板子如何做连板调试,我试了他的两个USB接口都不行。系统装的是Ubuntu 22.04。

--> Init runtime environment I target set by user is: rk3588 I Starting ntp or adb, target is RK3588 I Start adb... I Connect to Device success! I NPUTransfer: Starting...

![errlog](https://user-images.githubusercontent.com/54522189/198229486-cadd4000-3301-40a6-9c27-fb8d3a68f51d.png)

Hi, I am loading and building a tensorflow model middlebury_d400.pb and encounter the following error message: ``` --> config model done --> Loading tensorflow model W:tensorflow:From /home/paul/rknn2/lib/python3.6/site-packages/tensorflow/python/tools/strip_unused_lib.py:86: extract_sub_graph (from tensorflow.python.framework.graph_util_impl)...

我的理解是cpu会把数据传到npu,执行每一层算子. cpu刚开始就会生成整个npu的计算图,在算子执行前后不需要cpu参与计算吗

where is the folder named docker?

不可否认,现在rknn-toolkit2以及rknpu2组合,有很多问题,大多是在目标板子上运行时,出来结果和PC上不一致(有时甚至在同一个板子多次运行间不一致)。 希望可以提供必要的开源或者底层接口,这样可以调试bug,或者自己实现Op。 谢谢