Error In This Simple Code

1

I wanted to try out some simple ops in pytorch and wrote the following in a jupyter notebook:

conv = nn.Conv2d(in_channels=1, out_channels=4, kernel_size=4)

x = np.random.randn(1, 1, 32, 32)
x = torch.tensor(x)

output = conv(x)

Yet somehow this simple piece of code throws an error.
Expected object of type torch.DoubleTensor but found type torch.FloatTensor for argument #2 'weight'

Why is this happening and more importantly, should this be happening?

2

Numpy uses np.float64 by default. Try to specify the dtype as np.float32 while creating x.
Alternatively, cast your tensor to float.