PyTorch torch.linspace() method is “used to create a one-dimensional tensor of size steps whose values are evenly spaced from start to end, inclusive.”
From PyTorch 1.11, linspace requires the steps argument. Use steps=100 to restore the previous behavior.
torch.linspace(start, end, steps=100, out=None)
- start: The starting value for the set of points.
- end: The ending value for the set of points.
- steps: The gap between each pair of adjacent points. Default: 100.
- out(Tensor, optional): The output tensor.
Example 1: How to Use torch.linspace() method
import torch # Create a tensor of 10 points between 0 and 1 tensor = torch.linspace(0, 1, 10) print(tensor)
tensor([0.0000, 0.1111, 0.2222, 0.3333, 0.4444, 0.5556, 0.6667, 0.7778, 0.8889, 1.0000])
Example 2: Generating Negative Ranges
Generate a tensor with 5 points between -1 and -5.
import torch # Create a tensor of 5 points between -1 and -5 tensor1 = torch.linspace(-1, -5, 5) print(tensor1)
tensor([-1., -2., -3., -4., -5.])
Example 3: Using the requires_grad Parameter
Generate a tensor with 7 points between 2 and 10 and set requires_grad=True to be part of the computation with gradient tracking.
import torch # Create a tensor of 7 points between 2 and 10 # with gradient computation enabled tensor2 = torch.linspace(2, 10, 7, requires_grad=True) print(tensor2)
tensor([ 2.0000, 3.3333, 4.6667, 6.0000, 7.3333, 8.6667, 10.0000], requires_grad=True)