get_order_grouped

full name: tenpy.models.lattice.get_order_grouped
parent module: tenpy.models.lattice
type: function

tenpy.models.lattice.get_order_grouped(shape, groups, priority=None)[source]

Variant of get_order(), grouping some sites of the unit cell.

This function is useful for lattices with a unit cell of more than 2 sites (e.g. Kagome). For 2D lattices with a unit cell, the ordering goes first within a group , then along y, then the next group (for the same x-value), again along y, and finally along x when all groups are done.

As an example, consider the Kagome lattice.

(Source code, png, hires.png, pdf)

../_images/tenpy-models-lattice-get_order_grouped-1.png

Note

In this doc-string, the word ‘direction’ refers to a physical direction of the lattice or the index u of the unit cell as an “artificial direction”.

Parameters:

shape (tuple of int) – The shape of the lattice, i.e., the length in each direction.
groups (tuple of tuple of int) – A partition and reordering of range(shape[-1]) into smaller groups. The ordering goes first within a group, then along the last spatial dimensions, then changing between different groups and finally in Cstyle order along the remaining spatial dimensions.
priority (None | tuple of ints) – By default (None), use C-style order for everything except the unit cell, as shown above. If a tuple, it should have length len(shape) and specifies which order to go first, similarly as in get_order(). To group sites in the unit cell, you should make the last entry of priority the largest. However, you can also choose to group along another direction - in that case groups should be a partitioning of range(shape(argmax(priority))). Try and plot it, if you need it!

Returns:

order – An order of the sites for Lattice.order in the specified ordering.

Return type:

ndarray (np.prod(shape), len(shape))