I wanted to learn about the use of the up sampling API, so I built a matrix, and I sampled it up:
This is my code:
model=torch.nn.Sequential()
model.add_module('upsamling2d-1',torch.nn.Upsample(size=(4,4)))
Input=np.asarray([[1,2],[3,4]])
Input=Input.reshape(1,1,2,2)
Input=torch.from_numpy(Input)
print(Input)
yhat=model(Input)
yhat=yhat.reshape((4,4))
print(yhat)
And this is the error:
how should i handle it
OK，I found the solution:
Input=Input.float()
This is a solution, but why does pytorch force us to increase the memory foot print of an int\byte tensor when doing a simple nearest neighbour interpolation?
I guess the main reason is that this function is not (yet) implemented for int/byte types, since these types cannot be used for gradient calculation.
If you think this feature is not an edge case and would be used by a lot of users, we would be more than happy to accept contributions. 
I find it now works for torch.uint8 but not for torch.long . Is there a reason why torch only allows for short int but not the long one?
unit8 is the standard image format, which is why interpolation is supported on this integer format. My guess about the lack of long support would be that it might not have been requested yet (so wasn’t implemented) and would increase the binary size as a new dtype is added to the upsample methods (apparently for an edge case).
If your data isn’t overflowing in float, you could transform if before applying the interpolation.