PyTorch basic computation function song

Here’s the list of functions with their descriptions:

• torch.abs(input) – Returns the absolute value of each element in the input tensor.
• torch.ceil(input) – Returns the smallest integer greater than or equal to each element in the input tensor (ceiling function).
• torch.floor(input) – Returns the largest integer less than or equal to each element in the input tensor (floor function).
• torch.clamp(input, min=None, max=None) – Clamps all elements in the input tensor to the range [min, max].
• torch.std(input, unbiased=True) – Returns the standard deviation of all elements in the input tensor.
• torch.prod(input, dim=None) – Returns the product of all elements in the input tensor or along the specified dimension.
• torch.unique(input, sorted=False, return_inverse=False) – Returns the unique elements of the input tensor.

Examples:

1. torch.abs(input)

• Description: Returns the absolute value of each element in the input tensor.
• Example:
  python a = torch.tensor([-1.5, 2.3, -3.7]) result = torch.abs(a) print(result)
  Output:
  tensor([1.5000, 2.3000, 3.7000])

1. torch.ceil(input)

• Description: Returns the smallest integer greater than or equal to each element in the input tensor (ceiling function).
• Example:
  python a = torch.tensor([1.2, 2.8, -3.5]) result = torch.ceil(a) print(result)
  Output:
  tensor([ 2., 3., -3.])

1. torch.floor(input)

• Description: Returns the largest integer less than or equal to each element in the input tensor (floor function).
• Example:
  python a = torch.tensor([1.2, 2.8, -3.5]) result = torch.floor(a) print(result)
  Output:
  tensor([ 1., 2., -4.])

1. torch.clamp(input, min=None, max=None)

• Description: Clamps all elements in the input tensor into the range [min, max]. If min or max are not specified, they are ignored.
• Example:
  python a = torch.tensor([-1, 2, 3, 4, 5]) result = torch.clamp(a, min=0, max=4) print(result)
  Output:
  tensor([0, 2, 3, 4, 4])

1. torch.std(input, unbiased=True)

• Description: Returns the standard deviation of all elements in the input tensor. By default, it uses the unbiased estimator.
• Example:
  python a = torch.tensor([1.0, 2.0, 3.0, 4.0]) result = torch.std(a) print(result)
  Output:
  tensor(1.2909)

1. torch.prod(input, dim=None)

• Description: Returns the product of all elements in the input tensor along the specified dimension dim. If dim is not specified, it returns the product of all elements.
• Example:
  python a = torch.tensor([1, 2, 3, 4]) result = torch.prod(a) print(result)
  Output:
  tensor(24)

1. torch.unique(input, sorted=False, return_inverse=False)

• Description: Returns the unique elements of the input tensor. If sorted=True, the output is sorted. return_inverse=True returns an additional tensor showing the indices for reconstructing the input.
• Example:
  python a = torch.tensor([1, 2, 2, 3, 4, 4, 5]) result = torch.unique(a) print(result)
  Output:
  tensor([1, 2, 3, 4, 5])