rknn-toolkit
rknn-toolkit copied to clipboard
rockchip-linux
Hi all, I manage to get my custom RKNNs running on RK3588, but when I added the post-processing step, I got overlapped BBoxes on the whole output image, unlike the...
环境：rknn_toolkit-1.7.3-cp36-cp36m-win_amd64.whl 在PC上测试发现一个bug， verbose_file如果是自定义路径，则一切正常。 verbose_file如果是None，则报错 Traceback (most recent call last): File "D:\code\Anaconda3\envs\toy\lib\weakref.py", line 624, in _exitfunc f() File "D:\code\Anaconda3\envs\toy\lib\weakref.py", line 548, in __call__ return info.func(*info.args, **(info.kwargs or {})) File "D:\code\Anaconda3\envs\toy\lib\tempfile.py", line...
\root@db5694cd7562:/MASR/rknn_convert# python rknn_conversion_with_hidden.py ../lstm_1_1_67_80_ctc_161024_with_hidden.pt ******************** W Input channel 1 less then 3, please update NPU Driver to v1.3.0 or later. W Input channel 1 less then 3, please update NPU...
当我使用examples测试的时候 提示没有rknnlite 可我已经在rlnn-tolkit-lite2下的pakages安装了lite的wheel 我用的是1.5.0的
ERROR: Failed building wheel for onnx I find when I install rknn_toolkit-1.7.3, it will automatically install onnx 1.6.0，It raise error as above. when I checked the onnx version at: https://pypi.org/project/onnx/1.7.0/#files,...
在使用rknn-toolkit2 （版本1.4.0），onnx 转 rknn的 加载模型这一步，加载不成功，报错信息如下，请问怎么解决？ W __init__: rknn-toolkit2 version: 1.4.0-22dcfef4 --> Loading model W load_onnx: If you don't need to crop the model, don't set 'inputs'/'input_size_list'/'outputs'! E load_onnx: Catch exception...
按照RKNN_OP_Support_And_Limit.xlsx中的测试版本配置的环境，RKNN-Toolkit 1.7.3、Pytorch 1.10.0、ONNX opset-verison 12 转换ERROR如下： W Call onnx.optimizer.optimize fail, skip optimize 1.6.0 --> Config model done --> Loading model /home/chenchen/anaconda3/envs/rknn/lib/python3.6/site-packages/rknn/api/rknn.py:150: RuntimeWarning: overflow encountered in exp ret = self.rknn_base.load_onnx(model, inputs,...
