I have a tensor with the size of (batch_size= 16,channels= 192,H=7,W=7). I want to decrease the number of channels to 160 channels but having the same height and width. The my class is the main class that I use for running the experiment and the x (the mentioned tensor), is the input to the my class:
class MyConv2d(nn.Module):
def __init__(self, in_planes, out_planes, kernel_size, stride, padding=0):
super(MyConv2d, self).__init__()
self.conv = nn.Conv2d(in_planes, out_planes, kernel_size=kernel_size, stride=stride, padding=padding, bias=False)
self.bn = nn.BatchNorm2d(out_planes, eps=1e-3, momentum=0.001, affine=True)
self.relu = nn.ReLU()
def forward(self, x):
x = self.conv(x) # The error happens here
x = self.bn(x)
x = self.relu(x)
return x
class my(nn.Module):
def __init__(self, in_channel=192, out_channel=160, out_sigmoid=False):
super(my, self).__init__()
self.out_sigmoid=out_sigmoid
self.deconv = self._make_deconv(in_channel, out_channel)
def _make_deconv(self, in_channel, out_channel, kernel_size=3, stride=1, padding=1):
layers=[]
layers.append(MyConv2d(in_channel, out_channel ,kernel_size=kernel_size, stride=stride, padding=padding))
return nn.Sequential(*layers)
def forward(self, x):
x=self.deconv(x)
return x
Unfortunately I face with the following error:
How can I fix the problem?