通常，输入数据的blob维度为4D
如何在训练原型文件中设置大小blob？
输入数据以HDF5格式保存，包括batch x channel x temporal x height x weight
我得到的错误是
I0520 21:39:35.705278 30447 net.cpp:84] Creating Layer conv1
I0520 21:39:35.705286 30447 net.cpp:381] conv1 <- data
I0520 21:39:35.705294 30447 net.cpp:339] conv1 -> conv1
I0520 21:39:35.705307 30447 net.cpp:113] Setting up conv1
F0520 21:39:35.705329 30447 base_conv_layer.cpp:14] Check failed: 4 == bottom[0]->num_axes() (4 vs. 5) Input must have 4 axes, corresponding to (num, channels, height, width)
输入blob维度似乎应为4。
有没有解决方案？
谢谢。