Advanced API documentation and generated content#

This page contains general code elements that are common for package documentation.

Autosummary table and API stub pages#


Drop specified labels from rows or columns.

pandas.DataFrame.groupby([by, axis, level, ...])

Group DataFrame using a mapper or by a Series of columns.


The ExtensionArray of the data backing this Series or Index.

Inline module documentation#



The NumPy linear algebra functions rely on BLAS and LAPACK to provide efficient low level implementations of standard linear algebra algorithms. Those libraries may be provided by NumPy itself using C versions of a subset of their reference implementations but, when possible, highly optimized libraries that take advantage of specialized processor functionality are preferred. Examples of such libraries are OpenBLAS, MKL (TM), and ATLAS. Because those libraries are multithreaded and processor dependent, environmental variables and external packages such as threadpoolctl may be needed to control the number of threads or specify the processor architecture.

Please note that the most-used linear algebra functions in NumPy are present in the main numpy namespace rather than in numpy.linalg. There are: dot, vdot, inner, outer, matmul, tensordot, einsum, einsum_path and kron.

Functions present in numpy.linalg are listed below.

Matrix and vector products#

multi_dot matrix_power


cholesky qr svd

Matrix eigenvalues#

eig eigh eigvals eigvalsh

Norms and other numbers#

norm cond det matrix_rank slogdet

Solving equations and inverting matrices#

solve tensorsolve lstsq inv pinv tensorinv




Compute the eigenvalues and right eigenvectors of a square array.

a(…, M, M) array

Matrices for which the eigenvalues and right eigenvectors will be computed

w(…, M) array

The eigenvalues, each repeated according to its multiplicity. The eigenvalues are not necessarily ordered. The resulting array will be of complex type, unless the imaginary part is zero in which case it will be cast to a real type. When a is real the resulting eigenvalues will be real (0 imaginary part) or occur in conjugate pairs

v(…, M, M) array

The normalized (unit “length”) eigenvectors, such that the column v[:,i] is the eigenvector corresponding to the eigenvalue w[i].


If the eigenvalue computation does not converge.

See also


eigenvalues of a non-symmetric array.


eigenvalues and eigenvectors of a real symmetric or complex Hermitian (conjugate symmetric) array.


eigenvalues of a real symmetric or complex Hermitian (conjugate symmetric) array.


Similar function in SciPy that also solves the generalized eigenvalue problem.


Best choice for unitary and other non-Hermitian normal matrices.


New in version 1.8.0.

Broadcasting rules apply, see the numpy.linalg documentation for details.

This is implemented using the _geev LAPACK routines which compute the eigenvalues and eigenvectors of general square arrays.

The number w is an eigenvalue of a if there exists a vector v such that a @ v = w * v. Thus, the arrays a, w, and v satisfy the equations a @ v[:,i] = w[i] * v[:,i] for \(i \in \{0,...,M-1\}\).

The array v of eigenvectors may not be of maximum rank, that is, some of the columns may be linearly dependent, although round-off error may obscure that fact. If the eigenvalues are all different, then theoretically the eigenvectors are linearly independent and a can be diagonalized by a similarity transformation using v, i.e, inv(v) @ a @ v is diagonal.

For non-Hermitian normal matrices the SciPy function scipy.linalg.schur is preferred because the matrix v is guaranteed to be unitary, which is not the case when using eig. The Schur factorization produces an upper triangular matrix rather than a diagonal matrix, but for normal matrices only the diagonal of the upper triangular matrix is needed, the rest is roundoff error.

Finally, it is emphasized that v consists of the right (as in right-hand side) eigenvectors of a. A vector y satisfying y.T @ a = z * y.T for some number z is called a left eigenvector of a, and, in general, the left and right eigenvectors of a matrix are not necessarily the (perhaps conjugate) transposes of each other.


G. Strang, Linear Algebra and Its Applications, 2nd Ed., Orlando, FL, Academic Press, Inc., 1980, Various pp.


>>> from numpy import linalg as LA

(Almost) trivial example with real e-values and e-vectors.

>>> w, v = LA.eig(np.diag((1, 2, 3)))
>>> w; v
array([1., 2., 3.])
array([[1., 0., 0.],
       [0., 1., 0.],
       [0., 0., 1.]])

Real matrix possessing complex e-values and e-vectors; note that the e-values are complex conjugates of each other.

>>> w, v = LA.eig(np.array([[1, -1], [1, 1]]))
>>> w; v
array([1.+1.j, 1.-1.j])
array([[0.70710678+0.j        , 0.70710678-0.j        ],
       [0.        -0.70710678j, 0.        +0.70710678j]])

Complex-valued matrix with real e-values (but complex-valued e-vectors); note that a.conj().T == a, i.e., a is Hermitian.

>>> a = np.array([[1, 1j], [-1j, 1]])
>>> w, v = LA.eig(a)
>>> w; v
array([2.+0.j, 0.+0.j])
array([[ 0.        +0.70710678j,  0.70710678+0.j        ], # may vary
       [ 0.70710678+0.j        , -0.        +0.70710678j]])

Be careful about round-off error!

>>> a = np.array([[1 + 1e-9, 0], [0, 1 - 1e-9]])
>>> # Theor. e-values are 1 +/- 1e-9
>>> w, v = LA.eig(a)
>>> w; v
array([1., 1.])
array([[1., 0.],
       [0., 1.]])
numpy.linalg.matrix_power(a, n)#

Raise a square matrix to the (integer) power n.

For positive integers n, the power is computed by repeated matrix squarings and matrix multiplications. If n == 0, the identity matrix of the same shape as M is returned. If n < 0, the inverse is computed and then raised to the abs(n).


Stacks of object matrices are not currently supported.

a(…, M, M) array_like

Matrix to be “powered”.


The exponent can be any integer or long integer, positive, negative, or zero.

a**n(…, M, M) ndarray or matrix object

The return value is the same shape and type as M; if the exponent is positive or zero then the type of the elements is the same as those of M. If the exponent is negative the elements are floating-point.


For matrices that are not square or that (for negative powers) cannot be inverted numerically.


>>> from numpy.linalg import matrix_power
>>> i = np.array([[0, 1], [-1, 0]]) # matrix equiv. of the imaginary unit
>>> matrix_power(i, 3) # should = -i
array([[ 0, -1],
       [ 1,  0]])
>>> matrix_power(i, 0)
array([[1, 0],
       [0, 1]])
>>> matrix_power(i, -3) # should = 1/(-i) = i, but w/ f.p. elements
array([[ 0.,  1.],
       [-1.,  0.]])

Somewhat more sophisticated example

>>> q = np.zeros((4, 4))
>>> q[0:2, 0:2] = -i
>>> q[2:4, 2:4] = i
>>> q # one of the three quaternion units not equal to 1
array([[ 0., -1.,  0.,  0.],
       [ 1.,  0.,  0.,  0.],
       [ 0.,  0.,  0.,  1.],
       [ 0.,  0., -1.,  0.]])
>>> matrix_power(q, 2) # = -np.eye(4)
array([[-1.,  0.,  0.,  0.],
       [ 0., -1.,  0.,  0.],
       [ 0.,  0., -1.,  0.],
       [ 0.,  0.,  0., -1.]])
numpy.linalg.norm(x, ord=None, axis=None, keepdims=False)#

Matrix or vector norm.

This function is able to return one of eight different matrix norms, or one of an infinite number of vector norms (described below), depending on the value of the ord parameter.


Input array. If axis is None, x must be 1-D or 2-D, unless ord is None. If both axis and ord are None, the 2-norm of x.ravel will be returned.

ord{non-zero int, inf, -inf, ‘fro’, ‘nuc’}, optional

Order of the norm (see table under Notes). inf means numpy’s inf object. The default is None.

axis{None, int, 2-tuple of ints}, optional.

If axis is an integer, it specifies the axis of x along which to compute the vector norms. If axis is a 2-tuple, it specifies the axes that hold 2-D matrices, and the matrix norms of these matrices are computed. If axis is None then either a vector norm (when x is 1-D) or a matrix norm (when x is 2-D) is returned. The default is None.

New in version 1.8.0.

keepdimsbool, optional

If this is set to True, the axes which are normed over are left in the result as dimensions with size one. With this option the result will broadcast correctly against the original x.

New in version 1.10.0.

nfloat or ndarray

Norm of the matrix or vector(s).

See also


Similar function in SciPy.


For values of ord < 1, the result is, strictly speaking, not a mathematical ‘norm’, but it may still be useful for various numerical purposes.

The following norms can be calculated:


norm for matrices

norm for vectors


Frobenius norm



Frobenius norm


nuclear norm


max(sum(abs(x), axis=1))



min(sum(abs(x), axis=1))



sum(x != 0)


max(sum(abs(x), axis=0))

as below


min(sum(abs(x), axis=0))

as below


2-norm (largest sing. value)

as below


smallest singular value

as below



The Frobenius norm is given by [1]:

\(||A||_F = [\sum_{i,j} abs(a_{i,j})^2]^{1/2}\)

The nuclear norm is the sum of the singular values.

Both the Frobenius and nuclear norm orders are only defined for matrices and raise a ValueError when x.ndim != 2.



G. H. Golub and C. F. Van Loan, Matrix Computations, Baltimore, MD, Johns Hopkins University Press, 1985, pg. 15


>>> from numpy import linalg as LA
>>> a = np.arange(9) - 4
>>> a
array([-4, -3, -2, ...,  2,  3,  4])
>>> b = a.reshape((3, 3))
>>> b
array([[-4, -3, -2],
       [-1,  0,  1],
       [ 2,  3,  4]])
>>> LA.norm(a)
>>> LA.norm(b)
>>> LA.norm(b, 'fro')
>>> LA.norm(a, np.inf)
>>> LA.norm(b, np.inf)
>>> LA.norm(a, -np.inf)
>>> LA.norm(b, -np.inf)
>>> LA.norm(a, 1)
>>> LA.norm(b, 1)
>>> LA.norm(a, -1)
>>> LA.norm(b, -1)
>>> LA.norm(a, 2)
>>> LA.norm(b, 2)
>>> LA.norm(a, -2)
>>> LA.norm(b, -2)
1.8570331885190563e-016 # may vary
>>> LA.norm(a, 3)
5.8480354764257312 # may vary
>>> LA.norm(a, -3)

Using the axis argument to compute vector norms:

>>> c = np.array([[ 1, 2, 3],
...               [-1, 1, 4]])
>>> LA.norm(c, axis=0)
array([ 1.41421356,  2.23606798,  5.        ])
>>> LA.norm(c, axis=1)
array([ 3.74165739,  4.24264069])
>>> LA.norm(c, ord=1, axis=1)
array([ 6.,  6.])

Using the axis argument to compute matrix norms:

>>> m = np.arange(8).reshape(2,2,2)
>>> LA.norm(m, axis=(1,2))
array([  3.74165739,  11.22497216])
>>> LA.norm(m[0, :, :]), LA.norm(m[1, :, :])
(3.7416573867739413, 11.224972160321824)
numpy.linalg.tensorinv(a, ind=2)#

Compute the ‘inverse’ of an N-dimensional array.

The result is an inverse for a relative to the tensordot operation tensordot(a, b, ind), i. e., up to floating-point accuracy, tensordot(tensorinv(a), a, ind) is the “identity” tensor for the tensordot operation.


Tensor to ‘invert’. Its shape must be ‘square’, i. e., prod(a.shape[:ind]) == prod(a.shape[ind:]).

indint, optional

Number of first indices that are involved in the inverse sum. Must be a positive integer, default is 2.


a’s tensordot inverse, shape a.shape[ind:] + a.shape[:ind].


If a is singular or not ‘square’ (in the above sense).

See also

numpy.tensordot, tensorsolve


>>> a = np.eye(4*6)
>>> a.shape = (4, 6, 8, 3)
>>> ainv = np.linalg.tensorinv(a, ind=2)
>>> ainv.shape
(8, 3, 4, 6)
>>> b = np.random.randn(4, 6)
>>> np.allclose(np.tensordot(ainv, b), np.linalg.tensorsolve(a, b))
>>> a = np.eye(4*6)
>>> a.shape = (24, 8, 3)
>>> ainv = np.linalg.tensorinv(a, ind=1)
>>> ainv.shape
(8, 3, 24)
>>> b = np.random.randn(24)
>>> np.allclose(np.tensordot(ainv, b, 1), np.linalg.tensorsolve(a, b))

C++ API#

type MyType#

Some type

const MyType Foo(const MyType bar)#

Some function type thing

template<typename T, std::size_t N>
class std::array#

Some cpp class

float Sphinx::version#

The description of Sphinx::version.

int version#

The description of version.

typedef std::vector<int> List#

The description of List type.

enum MyEnum#

An unscoped enum.

enumerator A#
enum class MyScopedEnum#

A scoped enum.

enumerator B#
protected enum struct MyScopedVisibilityEnum : std::underlying_type<MySpecificEnum>::type#

A scoped enum with non-default visibility, and with a specified underlying type.

enumerator B#

JavaScript API#

class module_a.submodule.ModTopLevel()#
  • Link to ModTopLevel()

class module_b.submodule.ModNested()#

Generated Index#

Part of the sphinx build process in generate and index file: Index.

Optional parameter args#

At this point optional parameters cannot be generated from code. However, some projects will manually do it, like so:

This example comes from django-payments module docs.

class payments.dotpay.DotpayProvider(seller_id, pin[, channel=0[, lock=False], lang='pl'])#

This backend implements payments using a popular Polish gateway,

Due to API limitations there is no support for transferring purchased items.

  • seller_id – Seller ID assigned by Dotpay

  • pin – PIN assigned by Dotpay

  • channel – Default payment channel (consult reference guide)

  • lang – UI language

  • lock – Whether to disable channels other than the default selected above



Lorem ipsum dolor sit amet, consectetur adipiscing elit. Fusce congue elit eu hendrerit mattis.

Some data link Data_item_1.