finesse.plotting.graph module

Graph plotting.

finesse.plotting.graph.create_graphviz_table(fields, frequencies, homs)

finesse.plotting.graph.graphviz_draw(model=None, network=None, draw_labels=True, angle=0, overlap=True, ratio=0.45, edge_len=1.0, size=(13, 7), pad=(0.0, 0.0), format: Literal['png', 'svg'] = 'svg', maxiter=500, layout='neato', mode=None, path=None, show=True)

This should get merged with plot_graphviz at some point.

Draws a graphviz figure using neato layout. The default settings are tested to produce a passable drawing of the aLIGO DRMI graph.

Parameters

anglefloat or bool

The angle parameter rotates the graph by angle degrees relative to the first edge in the graph, which most of the time is the edge coming out of the laser. Set angle=False to disable rotation and let graphviz decide how to rotate the graph.

overlapbool or str

Setting for how graphviz deals with node overlaps. Set to False for graphviz to attempt to remove overlaps. Note that overlap removal runs as a post-processing step after initial layout and usually makes the graph look worse.

ratiofloat

Post processing step to stretch the graph. Used for stretching horizontally to compoensate for wider nodes to fit node labels.

pathPath or None

Save the resulting image to the given path. Defaults to None, which saves in a temporary file that is displayed if ‘show’ is set to True.

showbool, optional

Whether to show the resulting image. In Jupyter environments, shows the plot inline, otherwise opens a webbrowser for svgs and PIL for pngs. Defaults to True.

Notes

The svg format sometimes crops the image too hard, which results in clipped nodes or edges, if that happens increase the pad graph_attr or use the png format.

finesse.plotting.graph.graphviz_draw_beam_trace(model=None, network=None, draw_labels=True, angle=0, overlap=True, ratio=0.45, edge_len=1.0, size=(13, 7), pad=(0.5, 0.5), format: Literal['png', 'svg']='svg', maxiter=500, layout='neato', mode=None, cmap=<matplotlib.colors.ListedColormap object>)

finesse.plotting.graph.plot_dcfields_graph(model: Model, path: Path | str | None = None, show: bool = True, add_fields: bool = True, add_operators: bool = False, operator_labels: bool = False) → Path

Visualize the values of the DC fields in a model in a graph representation. Will run a DCFields / finesse.analysis.actions.dc.DCFields action under the hood.

Parameters

modelfinesse.Model

Model to visualize

pathPath | str | None, optional

Save the resulting image to the given path. Defaults to None, which saves in a temporary file that is displayed if ‘show’ is set to True.

showbool, optional

Whether to show the resulting image. In Jupyter environments, shows the plot inline, otherwise opens a webbrowser. Defaults to True.

add_fieldsbool, optional

Whether to show the DC fields values in the nodes of the graph, by default True

add_operatorsbool, optional

Whether to show the couplings between nodes as edge tooltips, by default False. Only supported for plane-wave single frequency models for now.

add_operatorsbool, optional

Show operator values directly as edge labels. Can clutter the view. By default False

Returns

pathlib.Path

Path where the svg was saved

Raises

ModuleNotFoundError

If pygraphviz is not installed

NotImplementedError

When add_operators is True, but the model is not plane-wave single frequency

finesse.plotting.graph.plot_graph(network, layout, graphviz=True, path=None, show=True, format: Literal['png', 'svg'] = 'svg', **kwargs)

finesse.plotting.graph.plot_graphviz(network, layout)

Plot graph with graphviz.

The pygraphviz Python package must be installed and available on the current Python path, and graphviz must be available on the system path.

Parameters

networknetworkx.Graph

The network to plot.

layoutstr

The layout type to use. Any layout algorithm provided by graphviz is supported.

Raises

ValueError

If the specified layout is not supported.

ImportError

If graphviz or pygraphviz is not installed.

finesse.plotting.graph.plot_nx_graph(network, layout, node_labels=True, node_attrs=False, edge_attrs=False, node_color_key=None, edge_color_key=None, label_font_size=12, attr_font_size=6, edge_font_size=6, bounding_ellipses=True, format: Literal['png', 'svg'] = 'svg', path: Path | None = None, show=True, **kwargs)

Plot graph with NetworkX.

Parameters

networknetworkx.Graph

The network to plot.

layoutstr

The layout type to use. Any layout algorithm provided by networkx.drawing.layout is supported.

node_labelsbool, optional

Show node names; defaults to True.

node_attrsbool or list, optional

Show node data. This can be True, in which case all node data is shown, or a list, in which case only the specified keys are shown. Defaults to True.

edge_attrsbool or list, optional

Show edge data. This can be True, in which case all edge data is shown, or a list, in which case only the specified keys are shown. Defaults to True.

node_color_keycallable, optional

Key function accepting a node and its attribute dict and returning a group. Each group is assigned a unique color. If not specified, nodes are not colored.

edge_color_keycallable, optional

Key function accepting an edge (u, v) and its attribute dict and returning a group. Each group is assigned a unique color. If not specified, edges are not colored.

label_font_size, attr_font_size, edge_font_sizeint, optional

Font size for node labels, attributes and edges. Defaults to 12, 6 and 6, respectively.

bounding_ellipses: bool, optional

Hijack the node label bounding boxes to draw the node labels inside of an ellipse (similar to graphviz neato layout). This guarantees the label is readable, but the arrow direction might not always be clear and might not combine well with node_color_key. Defaults to True.

pathPath or None

Save the resulting image to the given path. Defaults to None, which saves in a temporary file that is displayed if ‘show’ is set to True.

showbool, optional

Whether to show the resulting image. In Jupyter environments, shows the plot inline, otherwise opens a webbrowser for svgs and PIL for pngs. Defaults to True.

Other Parameters

kwargs

Anything else supported by networkx.drawing.nx_pylab.draw().

Raises

ValueError

If the specified layout is not supported.

Exception

If the graph cannot be represented with the specified layout.

finesse.plotting.graph.show_graphviz(A: pygraphviz.AGraph, format: str, path: str | Path | None, layout: str, show: bool) → pathlib.Path