lzlwakeup
lzlwakeup
I have same issue with Segmentation fault (core dumped). model in https://github.com/Linzaer/Ultra-Light-Fast-Generic-Face-Detector-1MB xxd convert onnx model to .h file, load model err in onnx_graph_alloc() in onnx.c line 1171, model g->nlen...
非常感谢!采用静态内存后,代码可以在dsp上进行初步的运行预测。
前面说的x_test我使用随机整形矩阵来解决。 后面报错是 (layer.weights[1])) IndexError: list index out of range 找到fuse部分,直接注释掉。 后面又报相同问题。 目前看起来是,conv2d每次需要写bias,但是部分模型conv2d+bn的组合,训练时直接bias=False,没有bias,导致程序运行中报错,因为程序默认是有bias的,我是这么认为的。 整理了一下,删除了之前的问题过程,现在修改的是没有bias的时候,layer.get_weights()长度1,set也是同样,所以在这个部分加判断,看是否需要bias。实际上目前用的模型,只要是符合融合conv+bn,都是没有bias的,考虑直接set(c_w),去掉bias。 有一处我在运行过程中有问题,就是说我最后一层是conv2d,直接输出,那么后面没有其他层了,这时 Traceback (most recent call last): File "gen_config.py", line 526, in print(layer.name, 'next', layer.outbound_nodes[0].outbound_layer.name) IndexError: list index...
I get xxd how to load input. Its my problem.
I have the same error win10 system
Speech signal processing requires overlap to process data, and the current processing requires the combination of a new frame and the previous frame. librosa.stft in python implements this process in...
Due to my mistake, the previous version of ort is consistent, which is not accurate. It should be added that the python version uses the win11 + torchcuda, and the...
Are there any new solutions or debugging methods?