jax.numpy.std#

jax.numpy.std(a,axis=None,dtype=None,out=None,ddof=0,keepdims=False,*,where=None,mean=None,correction=None)[source]#

Compute the standard deviation along a given axis.

JAX implementation ofnumpy.std().

Parameters:

• a (ArrayLike) – input array.

• axis (Axis) – optional, int or sequence of ints, default=None. Axis along which thestandard deviation is computed. If None, standard deviaiton is computedalong all the axes.

• dtype (DTypeLike |None) – The type of the output array. Default=None.

• ddof (int) – int, default=0. Degrees of freedom. The divisor in the standard deviationcomputation isN-ddof,N is number of elements along given axis.

• keepdims (bool) – bool, default=False. If true, reduced axes are left in the resultwith size 1.

• where (ArrayLike |None) – optional, boolean array, default=None. The elements to be used in thestandard deviation. Array should be broadcast compatible to the input.

• mean (ArrayLike |None) – optional, mean of the input array, computed along the given axis.If provided, it will be used to compute the standard deviation instead ofcomputing it from the input array. If specified, mean must be broadcast-compatiblewith the input array. In the general case, this can be achieved by computing the mean withkeepdims=True andaxis matching this function’saxis argument.

• correction (int |float |None) – int or float, default=None. Alternative name forddof.Both ddof and correction can’t be provided simultaneously.

• out (None) – Unused by JAX.

Returns:

An array of the standard deviation along the given axis.

Return type:

Array

See also

• jax.numpy.var(): Compute the variance of array elements over givenaxis.

• jax.numpy.mean(): Compute the mean of array elements over a given axis.

• jax.numpy.nanvar(): Compute the variance along a given axis, ignoringNaNs values.

• jax.numpy.nanstd(): Computed the standard deviation of a given axis,ignoring NaN values.

Examples

By default,jnp.std computes the standard deviation along all axes.

>>>x=jnp.array([[1,3,4,2],...[4,2,5,3],...[5,4,2,3]])>>>withjnp.printoptions(precision=2,suppress=True):...jnp.std(x)Array(1.21, dtype=float32)

Ifaxis=0, computes along axis 0.

>>>withjnp.printoptions(precision=2,suppress=True):...print(jnp.std(x,axis=0))[1.7  0.82 1.25 0.47]

To preserve the dimensions of input, you can setkeepdims=True.

>>>withjnp.printoptions(precision=2,suppress=True):...print(jnp.std(x,axis=0,keepdims=True))[[1.7  0.82 1.25 0.47]]

Ifddof=1:

>>>withjnp.printoptions(precision=2,suppress=True):...print(jnp.std(x,axis=0,keepdims=True,ddof=1))[[2.08 1.   1.53 0.58]]

To include specific elements of the array to compute standard deviation, youcan usewhere.

>>>where=jnp.array([[1,0,1,0],...[0,1,0,1],...[1,1,1,0]],dtype=bool)>>>jnp.std(x,axis=0,keepdims=True,where=where)Array([[2., 1., 1., 0.]], dtype=float32)