rknn-toolkit
rknn-toolkit copied to clipboard
rockchip-linux
各位，对pb模型使用rknn进行int8量化,如何对校准集里的每张图片减去各自的均值和方差，不用统一的均值方差，有什么办法吗
请问一下有人遇到过以下问题吗？已发帖子求助，感激不尽！！！ [python代码+USB摄像头目标检测报错](https://dev.t-firefly.com/thread-128479-2-1.html) 我在python3.8执行了python的目标检测demo，python3 rockx_object_detection.py，报新的错误AttributeError: /home/firefly/.local/lib/python3.8/site-packages/rockx/sdk/rockx-rk3399pro-Linux/lib64/librockx.so: undefined symbol: rockx_carplate_align_v2  我看网友遇到类似的问题，还没解决 [TB-3399ProD开发板,跑Rockx1.4.4异常](https://t.rock-chips.com/forum.php?mod=viewthread&tid=3643)
E Calc node Slice : Slice_27 output shape fail E Catch exception when loading onnx model: nanodet.onnx! E Traceback (most recent call last): E File "rknn/base/RKNNlib/onnx_ir/onnx_numpy_backend/shape_inference.py", line 65, in rknn.base.RKNNlib.onnx_ir.onnx_numpy_backend.shape_inference.infer_shape...
@rkchrome thank you for sharing so good product. I want to run rknn model in docker after converted .onnx to .rknn. Use load_rknn function, the total code shows as belows...
PC使用adb线与rk3399pro连接，将test.py中的ret = rknn.init_runtime()改为ret = rknn.init_runtime(target='rk3399pro',device_id='56c0d97c7b4de727')。运行test.py时报以下错误。 --> Init runtime environment I NPUTransfer: Starting NPU Transfer Client, Transfer version 2.1.0 (b5861e7@2020-11-23T11:50:36) E RKNNAPI: rknn_init, driver open fail! ret = -9(ERROR_PIPE)! E Catch...
[rknn-toolkit](https://github.com/rockchip-linux/rknn-toolkit)/[rknn-toolkit-lite](https://github.com/rockchip-linux/rknn-toolkit/tree/master/rknn-toolkit-lite)/[packages](https://github.com/rockchip-linux/rknn-toolkit/tree/master/rknn-toolkit-lite/packages)/rknn_toolkit_lite-1.7.3-cp37-cp37m-linux_armv7l.whl 该包在rv1126下debian10系统不可用  经过测试该连接下的whl可以使用：https://gitlab.com/firefly-linux/external/rknn-toolkit/-/tree/rv1126_rv1109/firefly/rknn-toolkit-lite/rknn-toolkit-lite-v1.7.0.dev_0cfb22/packages
Error message: ``` $ pip install rknn_toolkit-1.7.3-cp38-cp38-linux_x86_64.whl Looking in indexes: https://pypi.org/simple, https://pypi.ngc.nvidia.com ERROR: rknn_toolkit-1.7.3-cp38-cp38-linux_x86_64.whl is not a supported wheel on this platform. ``` System info: ubuntu 22.04 python 3.8 intel...
PyTorch 版本：2.0.0 rknn-toolkit 版本：v1.7.3和v1.7.5 请问为什么qat训练的模型转到rknn上还会有很大的损失？
报错信息如下： W:tensorflow:From E:\tools\anaconda\Anaconda_envs\envs\py36\lib\site-packages\rknn\api\rknn.py:150: The name tf.enable_eager_execution is deprecated. Please use tf.compat.v1.enable_eager_execution instead. 2024-07-17 11:05:14.298500: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2...
