Hi there,
I am a new of Pytorch.
I want to modify tensor x to y as a following rule:
y[i][j] = 1.0 ,if x[I][j] > 0,
y[i][j] = 0.0 ,if x[I][j] = 0,
y[i][j] = -1.0 ,if x[I][j] < 0.
For example:
x = tensor([
[0.0, 1.5, 0.5, 1.0],
[0.0, 0.0, 1.0, 1.5],
[-0.5, -1.0, -1.5, -0.5]
])
y = tensor([
[0.0, 1.0, 1.0, 1.0],
[0.0, 0.0, 1.0, 1.0],
[-1.0, -1.0, -1.0, -0.0]
])
(x should have -1.5, -1.0, -0.5, 0, 0.5, 1 or 1.5)
For now, I implemented them below code:
x[x>0] = 1.0
x[x<0] = -1.0
y = x
I am using GPUs to run the above code, but actually x and y are very large tensors so it takes a lot of time.
I guess it is because my code isn’t parallelized properly.
Does anyone have a better idea to modify the tensor?