|
你好,我在转换darknet版mobilenet出现错误,麻烦看看,报错信息:
T Traceback (most recent call last):
T File "rknn/api/rknn_base.py", line 381, in rknn.api.rknn_base.RKNNBase.load_darknet
T File "rknn/base/RKNNlib/converter/convert_darknet.py", line 471, in rknn.base.RKNNlib.converter.convert_darknet.Converter.load_blobs
T File "rknn/base/RKNNlib/converter/convert_darknet.py", line 215, in rknn.base.RKNNlib.converter.convert_darknet.Convolution_Blob_Assign.assign
T File "<__array_function__ internals>", line 6, in reshape
T File "/home/mike/anaconda3/envs/rk1808/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 301, in reshape
T return _wrapfunc(a, 'reshape', newshape, order=order)
T File "/home/mike/anaconda3/envs/rk1808/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 61, in _wrapfunc
T return bound(*args, **kwds)
T ValueError: cannot reshape array of size 40607 into shape (384,384,3,3)
在模型的cfg中深度可分离卷积是这样定义的:
[convolutional]
groups=96
filters=96
size=3
pad=1
stride=2
batch_normalize=1
activation=relu
谢谢~
|
|