_reduce_std

reduce_std

Compute the std value along a given axis of a tensor.

This method can be used with Numpy data:

n = np.array([[[1.0, 2.0], [3.0, 4.0]], [[5.0, 6.0], [7.0, 8.0]]])
b = fe.backend.reduce_std(n)  # 2.2913
b = fe.backend.reduce_std(n, axis=0)  # [[2., 2.], [2., 2.]]
b = fe.backend.reduce_std(n, axis=1)  # [[1., 1.], [1., 1.]]
b = fe.backend.reduce_std(n, axis=[0,2])  # [2.23606798 2.23606798]

This method can be used with TensorFlow tensors:

t = tf.constant([[[1.0, 2.0], [3.0, 4.0]], [[5.0, 6.0], [7.0, 8.0]]])
b = fe.backend.reduce_std(t)  # 2.2913
b = fe.backend.reduce_std(t, axis=0)  # [[2., 2.], [2., 2.]]
b = fe.backend.reduce_std(t, axis=1)  # [[2, 3], [3, 7]]
b = fe.backend.reduce_std(t, axis=[0,2])  # [2.23606798 2.23606798]

This method can be used with PyTorch tensors:

p = torch.tensor([[[1.0, 2.0], [3.0, 4.0]], [[5.0, 6.0], [7.0, 8.0]]])
b = fe.backend.reduce_std(p)  # 2.2913
b = fe.backend.reduce_std(p, axis=0)  # [[2., 2.], [2., 2.]]
b = fe.backend.reduce_std(p, axis=1)  # [[1., 1.], [1., 1.]]
b = fe.backend.reduce_std(p, axis=[0,2])  # [2.23606798 2.23606798]

Parameters:

Name	Type	Description	Default
tensor	Tensor	The input value.	required
axis	Union[None, int, Sequence[int]]	Which axis or collection of axes to compute the std along.	None
keepdims	bool	Whether to preserve the number of dimensions during the reduction.	False

Returns:

Type	Description
Tensor	The std values of tensor along axis.

Raises:

Type	Description
ValueError	If tensor is an unacceptable data type.

Source code in fastestimator/fastestimator/backend/_reduce_std.py

def reduce_std(tensor: Tensor, axis: Union[None, int, Sequence[int]] = None, keepdims: bool = False) -> Tensor:
    """Compute the std value along a given `axis` of a `tensor`.

    This method can be used with Numpy data:
    ```python
    n = np.array([[[1.0, 2.0], [3.0, 4.0]], [[5.0, 6.0], [7.0, 8.0]]])
    b = fe.backend.reduce_std(n)  # 2.2913
    b = fe.backend.reduce_std(n, axis=0)  # [[2., 2.], [2., 2.]]
    b = fe.backend.reduce_std(n, axis=1)  # [[1., 1.], [1., 1.]]
    b = fe.backend.reduce_std(n, axis=[0,2])  # [2.23606798 2.23606798]
    ```

    This method can be used with TensorFlow tensors:
    ```python
    t = tf.constant([[[1.0, 2.0], [3.0, 4.0]], [[5.0, 6.0], [7.0, 8.0]]])
    b = fe.backend.reduce_std(t)  # 2.2913
    b = fe.backend.reduce_std(t, axis=0)  # [[2., 2.], [2., 2.]]
    b = fe.backend.reduce_std(t, axis=1)  # [[2, 3], [3, 7]]
    b = fe.backend.reduce_std(t, axis=[0,2])  # [2.23606798 2.23606798]
    ```

    This method can be used with PyTorch tensors:
    ```python
    p = torch.tensor([[[1.0, 2.0], [3.0, 4.0]], [[5.0, 6.0], [7.0, 8.0]]])
    b = fe.backend.reduce_std(p)  # 2.2913
    b = fe.backend.reduce_std(p, axis=0)  # [[2., 2.], [2., 2.]]
    b = fe.backend.reduce_std(p, axis=1)  # [[1., 1.], [1., 1.]]
    b = fe.backend.reduce_std(p, axis=[0,2])  # [2.23606798 2.23606798]
    ```

    Args:
        tensor: The input value.
        axis: Which axis or collection of axes to compute the std along.
        keepdims: Whether to preserve the number of dimensions during the reduction.

    Returns:
        The std values of `tensor` along `axis`.

    Raises:
        ValueError: If `tensor` is an unacceptable data type.
    """
    if tf.is_tensor(tensor):
        return tf.math.reduce_std(tensor, axis=axis, keepdims=keepdims)
    elif isinstance(tensor, torch.Tensor):
        if axis is None:
            if not keepdims:
                return tensor.std(unbiased=False)
            axis = list(range(len(tensor.shape)))
        tensor = tensor.std(dim=axis, unbiased=False, keepdim=keepdims)
        return tensor
    elif isinstance(tensor, np.ndarray):
        if isinstance(axis, list):
            axis = tuple(axis)
        return np.std(tensor, axis=axis, keepdims=keepdims)
    else:
        raise ValueError("Unrecognized tensor type {}".format(type(tensor)))