conv2d的输入_Conv1d, conv2d, conv3d
conv1d
输入数据格式为(batch_size, channel, length)
nn.conv1d(in_channels, outchannel, kernel_size, stride=1,padding=0,dilation=1,groups=1)
nn.Conv1d(1, 20, 5)表示输入1通道，20个卷积核，核大小为5
conv2d
输入数据格式为(batch_size, channel, Height, Width),
nn.conv2d(in_channels,out_channels,kernel_size, stride=(1,1),padding=0,dilation=(1,1),groups=1)
nn.Conv2d(1, 20, (3, 3), stride=(1, 1),padding=(2,2)) 表示输入1通道，20个卷积核，核大小为(3*3)
针对conv2d , 输入的是4维，[150，103，7，7]
conv3d
输入数据格式为(batch_size, channel, Depth, Height, Width)
nn.conv2d(in_channels,out_channels,kernel_size, stride=(1,1,1),padding=0,dilation=(1,1,1),groups=1)
nn.Conv3d(1, 90, (24, 3, 3), stride=(9,1,1),padding=(1,1,1))