MatrixPotential2S¶

About the `MatrixPotential2S` class¶

The WaveBlocks Project

Inheritance diagram¶

Class documentation¶

class WaveBlocksND.MatrixPotential2S(expression, variables, **kwargs)[source]¶

This class represents a matrix potential $V(x)$ . The potential is given as an analytic $2 \times 2$ matrix expression. Some symbolic calculations with the potential are supported.

calculate_eigenvalues()[source]¶: Calculate the two eigenvalues $\lambda_i(x)$ of the potential $V(x)$ . We can do this by symbolic calculations. The multiplicities are taken into account. Note: This function is idempotent and the eigenvalues are memoized for later reuse.

calculate_eigenvectors()[source]¶: Calculate the two eigenvectors $\nu_i(x)$ of the potential $V(x)$ . We can do this by symbolic calculations. Note: This function is idempotent and the eigenvectors are memoized for later reuse.

calculate_exponential(factor=1)[source]¶

Calculate the matrix exponential $\exp(\alpha V)$ . In the case of this class the matrix is of size $2 \times 2$ thus the exponential can be calculated analytically for a general matrix. Note: This function is idempotent.

Parameters:	factor – The prefactor $\alpha$ in the exponential.

calculate_hessian()[source]¶: Calculate the Hessian matrix $\nabla^2 \lambda_i$ of the potential’s eigenvalues $\Lambda(x)$ with $x \in \mathbb{R}^D$ . For potentials which depend only one variable, this equals the second derivative and $D=1$ . Note that this function is idempotent.

calculate_jacobian()[source]¶: Calculate the Jacobian matrix $\nabla \lambda_i$ of the potential’s eigenvalues $\Lambda(x)$ with $x \in \mathbb{R}^D$ . For potentials which depend only one variable, this equals the first derivative and $D=1$ . Note that this function is idempotent.

calculate_local_quadratic(diagonal_component=None)[source]¶

Calculate the local quadratic approximation matrix $U(x)$ of the potential’s eigenvalues in $\Lambda(x)$ . This function can be used for the homogeneous case and takes into account the leading component $\chi \in [0,\ldots,N-1]$ . If the parameter $i$ is not given, calculate the local quadratic approximation matrix $U(x)$ of all the potential’s eigenvalues in $\Lambda$ . This case can be used for the inhomogeneous case.

Parameters:	diagonal_component (Integer or `None` (default)) – Specifies the index $i$ of the eigenvalue $\lambda_i$ that gets expanded into a Taylor series $u_i$ .

calculate_local_remainder(diagonal_component=None)[source]¶

Calculate the non-quadratic remainder matrix $W(x) = V(x) - U(x)$ of the quadratic approximation matrix $U(x)$ of the potential’s eigenvalue matrix $\Lambda(x)$ . In the homogeneous case the matrix $U$ is given by $U(x) = \text{diag}([u_i,\ldots,u_i])$ where in the inhomogeneous case it is given by $U(x) = \text{diag}([u_0,\ldots,u_{N-1}])$ .

Parameters:	diagonal_component (Integer or `None` (default)) – Specifies the index $i$ of the eigenvalue $\lambda_i$ that gets expanded into a Taylor series $u_i$ . If set to `None` the inhomogeneous case is computed.

evaluate_at(grid, entry=None, as_matrix=True)[source]¶

Evaluate the potential $V(x)$ elementwise on a grid $\Gamma$ .

Parameters:	grid (A `Grid` instance. (Numpy arrays are not directly supported yet.)) – The grid containing the nodes $\gamma_i$ we want to evaluate the potential at. entry (A python tuple of two integers.) – The indices $(i,j)$ of the component $V_{i,j}(x)$ we want to evaluate or `None` to evaluate all entries. as_matrix – Dummy parameter which has no effect here.
Returns:	A list containing 4 numpy ndarrays of shape $(1, \|\Gamma\|)$ .

evaluate_eigenvalues_at(grid, entry=None, as_matrix=False)[source]¶

Evaluate the eigenvalues $\lambda_i(x)$ elementwise on a grid $\Gamma$ .

Parameters:

grid (A Grid instance. (Numpy arrays are not directly supported yet.)) – The grid $\Gamma$ containing the nodes $\gamma_i$ we want to evaluate the eigenvalues at.
entry (A python tuple of two integers.) – The indices $(i,j)$ of the component $\Lambda_{i,j}(x)$ we want to evaluate or None to evaluate all entries. If $j = i$ then we evaluate the eigenvalue $\lambda_i(x)$ .
as_matrix – Whether to include the off-diagonal zero entries of $\Lambda_{i,j}(x)$ in the return value.

Returns:

A list containing the numpy ndarray, all of shape $(1, |\Gamma|)$ .

evaluate_eigenvectors_at(grid, entry=None)[source]¶

Evaluate the two eigenvectors $\nu_i(x)$ elementwise on a grid $\Gamma$ .

Parameters:	grid (A `Grid` instance. (Numpy arrays are not directly supported yet.)) – The grid containing the nodes $\gamma_i$ we want to evaluate the eigenvectors at. entry (A single python integer.) – The index $i$ of the eigenvector $\nu_i(x)$ we want to evaluate or `None` to evaluate all eigenvectors.
Returns:	A list containing the numpy ndarrays, all of shape $(N, \|\Gamma\|)$ .

evaluate_exponential_at(grid)[source]¶

Evaluate the exponential of the potential matrix $V(x)$ on a grid $\Gamma$ .

Parameters:	grid (A `Grid` instance. (Numpy arrays are not directly supported yet.)) – The grid containing the nodes $\gamma_i$ we want to evaluate the exponential at.
Returns:	The numerical approximation of the matrix exponential at the given grid nodes. A list contains the exponentials for all entries $(i,j)$ , each having the same shape as the grid.

evaluate_hessian_at(grid, component=None)[source]¶

Evaluate the list of Hessian matrices $\nabla^2 \lambda_i(x)$ at some grid nodes $\Gamma$ .

Parameters:	grid (A `Grid` instance. (Numpy arrays are not directly supported yet.)) – The grid nodes $\Gamma$ the Hessian gets evaluated at. component – Dummy parameter that has no effect here.
Returns:	The value of the potential’s Hessian at the given nodes. The result is an `ndarray` of shape $(D,D)$ is we evaluate at a single grid node or of shape $(\|\Gamma\|,D,D)$ if we evaluate at multiple nodes simultaneously.

evaluate_jacobian_at(grid, component=None)[source]¶

Evaluate the list of Jacobian matrices $\nabla \lambda_i(x)$ at some grid nodes $\Gamma$ .

Parameters:	grid (A `Grid` instance. (Numpy arrays are not directly supported yet.)) – The grid nodes $\Gamma$ the Jacobian gets evaluated at. component – Dummy parameter that has no effect here.
Returns:	The value of the potential’s Jacobian at the given nodes. The result is a list of `ndarray` each of shape $(D,1)$ is we evaluate at a single grid node or of shape $(D,\|\Gamma\|)$ if we evaluate at multiple nodes simultaneously.

evaluate_local_quadratic_at(grid, diagonal_component=None)[source]¶

Numerically evaluate the local quadratic approximation matrix $U(x)$ of the potential’s eigenvalues in $\Lambda(x)$ at the given grid nodes $\Gamma$ .

Parameters:	grid (A `Grid` instance. (Numpy arrays are not directly supported yet.)) – The grid $\Gamma$ containing the nodes $\gamma$ we want to evaluate the quadratic approximation at. diagonal_component – Specifies the index $i$ of the eigenvalue $\lambda_i$ that gets expanded into a Taylor series $u_i$ .
Returns:	A list of tuples or a single tuple. Each tuple $(\lambda, J, H)$ contains the the evaluated eigenvalues $\lambda_i(\Gamma)$ , the Jacobian $J(\Gamma)$ and the Hessian $H(\Gamma)$ in this order.

evaluate_local_remainder_at(grid, position, diagonal_component=None, entry=None)[source]¶

Numerically evaluate the non-quadratic remainder $W(x)$ of the quadratic approximation $U(x)$ of the potential’s eigenvalue $\Lambda(x)$ at the given nodes $\Gamma$ .

Warning: do not set the diagonal_component and the entry parameter both to None.

Parameters:

grid – The grid nodes $\Gamma$ the remainder $W$ gets evaluated at.
position – The point $q \in \mathbb{R}^D$ where the Taylor series is computed.
diagonal_component (Integer or None (default)) – Specifies the index $i$ of the eigenvalue $\lambda_i$ that gets expanded into a Taylor series $u_i$ and whose remainder matrix $W(x) = V(x) - \text{diag}([u_i,\ldots,u_i])$ we evaluate. If set to None the inhomogeneous case given by $W(x) = V(x) - \text{diag}([u_0,\ldots,u_{N-1}])$ is computed.
entry (A python tuple of two integers.) – The entry $\left(i,j\right)$ of the remainder matrix $W$ that is evaluated.

Returns:

A list with $N^2$ ndarray elements or a single ndarray. Each containing the values of $W_{i,j}(\Gamma)$ . Each array is of shape $(1,|\Gamma|)$ .

get_dimension()¶: Return the dimension $D$ of the potential $V(x)$ . The dimension is equal to the number of free variables $x_i$ where $x := (x_1, x_2, ..., x_D)$ .

get_number_components()¶: Return the number $N$ of components the potential $V(x)$ supports. This is equivalent to the number of energy levels $\lambda_i(x)$ .

MatrixPotential2S¶

About the `MatrixPotential2S` class¶

Inheritance diagram¶

Class documentation¶

Table Of Contents

Previous topic

Next topic

This Page

MatrixPotential2S¶

About the MatrixPotential2S class¶

Inheritance diagram¶

Class documentation¶

About the `MatrixPotential2S` class¶