network_plots

Network plot classes.

NetworkPlot

Bases: pathpyG.visualisations.plot.PathPyPlot

Network plot class for a static network.

Source code in src/pathpyG/visualisations/network_plots.py

class NetworkPlot(PathPyPlot):
    """Network plot class for a static network."""

    _kind = "network"

    def __init__(self, network: Graph, **kwargs: Any) -> None:
        """Initialize network plot class."""
        super().__init__()
        self.network = network
        self.config = kwargs
        self.generate()

    def generate(self) -> None:
        """Generate the plot."""
        self._compute_edge_data()
        self._compute_node_data()
        self._compute_layout()
        self._cleanup_config()
        self._cleanup_data()

    def _compute_node_data(self) -> None:
        """Generate the data structure for the nodes."""
        # initialize values
        nodes: dict = {}
        attributes: set = {"color", "size", "opacity", "label"}
        attr: defaultdict = defaultdict(dict)

        # get attributes categories from pathpyg
        categories = {a.replace("node_", "") for a in self.network.node_attrs()}.intersection(attributes)

        # add node data to data dict
        self._get_node_data(nodes, attributes, attr, categories)

        # convert needed attributes to useful values
        attr["color"] = self._convert_color(attr["color"], mode="node")
        attr["opacity"] = self._convert_opacity(attr["opacity"], mode="node")
        attr["size"] = self._convert_size(attr["size"], mode="node")
        attr["label"] = self._convert_label(attr["label"], mode="node")

        # update data dict with converted attributes
        for attribute in attr:
            for key, value in attr[attribute].items():
                nodes[key][attribute] = value

        # save node data
        self.data["nodes"] = nodes

    def _get_node_data(
        self,
        nodes: dict,
        attributes: set,
        attr: defaultdict,
        categories: set,
    ) -> None:
        """Extract node data from network."""
        for uid in self.network.nodes:
            nodes[uid] = {"uid": str(uid)}

            # add edge attributes if needed
            for attribute in attributes:
                attr[attribute][uid] = (
                    self.network[f"node_{attribute}", uid].item() if attribute in categories else None
                )

    def _compute_edge_data(self) -> None:
        """Generate the data structure for the edges."""
        # initialize values
        edges: dict = {}
        attributes: set = {"weight", "color", "size", "opacity"}
        attr: defaultdict = defaultdict(dict)

        # get attributes categories from pathpyg
        categories: set = {a.replace("edge_", "") for a in self.network.edge_attrs()}.intersection(attributes)

        # add edge data to data dict
        self._get_edge_data(edges, attributes, attr, categories)

        # convert needed attributes to useful values
        attr["weight"] = self._convert_weight(attr["weight"], mode="edge")
        attr["color"] = self._convert_color(attr["color"], mode="edge")
        attr["opacity"] = self._convert_opacity(attr["opacity"], mode="edge")
        attr["size"] = self._convert_size(attr["size"], mode="edge")

        # update data dict with converted attributes
        for attribute in attr:
            for key, value in attr[attribute].items():
                edges[key][attribute] = value

        # save edge data
        self.data["edges"] = edges

    def _get_edge_data(
        self,
        edges: dict,
        attributes: set,
        attr: defaultdict,
        categories: set,
    ) -> None:
        """Extract edge data from network."""
        for u, v in self.network.edges:
            uid = f"{u}-{v}"
            edges[uid] = {
                "uid": uid,
                "source": str(u),
                "target": str(v),
            }
            # add edge attributes if needed
            for attribute in attributes:
                attr[attribute][uid] = (
                    self.network[f"edge_{attribute}", u, v].item() if attribute in categories else None
                )

    def _convert_weight(self, weight: dict, mode: str = "node") -> dict:
        """Convert weight to float."""
        # get style from the config
        style = self.config.get(f"{mode}_weight")

        # check if new attribute is a single object
        if isinstance(style, (int, float)):
            weight = {k: style for k in weight}

        # check if new attribute is a dict
        elif isinstance(style, dict):
            weight.update(**{k: v for k, v in style.items() if k in weight})

        # return all weights which are not None
        return {k: v if v is not None else 1 for k, v in weight.items()}

    def _convert_color(self, color: dict, mode: str = "node") -> dict:
        """Convert colors to hex if rgb."""
        # get style from the config
        style = self.config.get(f"{mode}_color")

        # check if new attribute is a single object
        if isinstance(style, (str, int, float, tuple)):
            color = {k: style for k in color}

        # check if new attribute is a dict
        elif isinstance(style, dict):
            color.update(**{k: v for k, v in style.items() if k in color})

        # check if new attribute is a list
        elif isinstance(style, list):
            for i, k in enumerate(color):
                try:
                    color[k] = style[i]
                except IndexError:
                    pass

        # check if numerical values are given
        values = [v for v in color.values() if isinstance(v, (int, float))]

        if values:
            # load colormap to map numerical values to color
            cmap = self.config.get(f"{mode}_cmap", Colormap())
            cdict = {values[i]: tuple(c[:3]) for i, c in enumerate(cmap(values, bytes=True))}

        # convert colors to hex if not already string
        for key, value in color.items():
            if isinstance(value, tuple):
                color[key] = rgb_to_hex(value)
            elif isinstance(value, (int, float)):
                color[key] = rgb_to_hex(cdict[value])

        # return all colors wich are not None
        return {k: v for k, v in color.items() if v is not None}

    def _convert_opacity(self, opacity: dict, mode: str = "node") -> dict:
        """Convert opacity to float."""
        # get style from the config
        style = self.config.get(f"{mode}_opacity")

        # check if new attribute is a single object
        if isinstance(style, (int, float)):
            opacity = {k: style for k in opacity}

        # check if new attribute is a dict
        elif isinstance(style, dict):
            opacity.update(**{k: v for k, v in style.items() if k in opacity})

        # return all colors wich are not None
        return {k: v for k, v in opacity.items() if v is not None}

    def _convert_size(self, size: dict, mode: str = "node") -> dict:
        """Convert size to float."""
        # get style from the config
        style = self.config.get(f"{mode}_size")

        # check if new attribute is a single object
        if isinstance(style, (int, float)):
            size = {k: style for k in size}

        # check if new attribute is a dict
        elif isinstance(style, dict):
            size.update(**{k: v for k, v in style.items() if k in size})

        # return all colors wich are not None
        return {k: v for k, v in size.items() if v is not None}

    def _convert_label(self, label: dict, mode: str = "node") -> dict:
        """Convert label to string."""
        # get style from the config
        style = self.config.get(f"{mode}_label")

        # check if new attribute is a single object
        if isinstance(style, str):
            label = {k: style for k in label}

        # check if new attribute is a dict
        elif isinstance(style, dict):
            label.update(**{k: v for k, v in style.items() if k in label})

        # check if new attribute is a list
        elif isinstance(style, list):
            for i, k in enumerate(label):
                try:
                    label[k] = style[i]
                except IndexError:
                    pass

        # return all labels wich are not None
        return {k: v for k, v in label.items() if v is not None}

    def _compute_layout(self) -> None:
        """Create layout."""
        # get layout form the config
        layout = self.config.get("layout", "rand")

        # if no layout is considered stop this process
        if layout is None:
            return

        # get layout dict for each node
        if isinstance(layout, str):
            layout = network_layout(self.network, layout=layout)
        elif not isinstance(layout, dict):
            logger.error("The provided layout is not valid!")
            raise AttributeError

        # update x,y position of the nodes
        for uid, (_x, _y) in layout.items():
            self.data["nodes"][uid]["x"] = _x
            self.data["nodes"][uid]["y"] = _y

    def _cleanup_config(self) -> None:
        """Clean up final config file."""
        try:
            directed = self.network.is_directed()
        except NotImplementedError:
            directed = False

        if not self.config.get("directed", None):
            self.config["directed"] = directed

        if not self.config.get("curved", None):
            self.config["curved"] = directed

    def _cleanup_data(self) -> None:
        """Clean up final data structure."""
        self.data["nodes"] = list(self.data["nodes"].values())
        self.data["edges"] = list(self.data["edges"].values())

__init__

Initialize network plot class.

Source code in src/pathpyG/visualisations/network_plots.py

def __init__(self, network: Graph, **kwargs: Any) -> None:
    """Initialize network plot class."""
    super().__init__()
    self.network = network
    self.config = kwargs
    self.generate()

generate

Generate the plot.

Source code in src/pathpyG/visualisations/network_plots.py

def generate(self) -> None:
    """Generate the plot."""
    self._compute_edge_data()
    self._compute_node_data()
    self._compute_layout()
    self._cleanup_config()
    self._cleanup_data()

StaticNetworkPlot

Bases: pathpyG.visualisations.network_plots.NetworkPlot

Network plot class for a static network.

Source code in src/pathpyG/visualisations/network_plots.py

class StaticNetworkPlot(NetworkPlot):
    """Network plot class for a static network."""

    _kind = "static"

TemporalNetworkPlot

Bases: pathpyG.visualisations.network_plots.NetworkPlot

Network plot class for a temporal network.

Source code in src/pathpyG/visualisations/network_plots.py

class TemporalNetworkPlot(NetworkPlot):
    """Network plot class for a temporal network."""

    _kind = "temporal"

    def __init__(self, network: TemporalGraph, **kwargs: Any) -> None:
        """Initialize network plot class."""
        super().__init__(network, **kwargs)

    def _get_edge_data(self, edges: dict, attributes: set, attr: defaultdict, categories: set) -> None:
        """Extract edge data from temporal network."""
        # TODO: Fix typing issue with temporal graphs
        for u, v, t in self.network.temporal_edges:  # type: ignore
            uid = f"{u}-{v}-{t}"
            edges[uid] = {
                "uid": uid,
                "source": str(u),
                "target": str(v),
                "start": int(t),
                "end": int(t) + 1,
            }
            # add edge attributes if needed
            for attribute in attributes:
                attr[attribute][uid] = (
                    self.network[f"edge_{attribute}", u, v].item() if attribute in categories else None
                )

    def _get_node_data(self, nodes: dict, attributes: set, attr: defaultdict, categories: set) -> None:
        """Extract node data from temporal network."""

        time = {e[2] for e in self.network.temporal_edges}

        if self.config.get("end", None) is None:
            self.config["end"] = int(max(time) + 1)

        if self.config.get("start", None) is None:
            self.config["start"] = int(min(time) - 1)

        for uid in self.network.nodes:
            nodes[uid] = {
                "uid": uid,
                "start": int(min(time) - 1),
                "end": int(max(time) + 1),
            }

            # add edge attributes if needed
            for attribute in attributes:
                attr[attribute][uid] = (
                    self.network[f"node_{attribute}", uid].item() if attribute in categories else None
                )

__init__

Initialize network plot class.

Source code in src/pathpyG/visualisations/network_plots.py

def __init__(self, network: TemporalGraph, **kwargs: Any) -> None:
    """Initialize network plot class."""
    super().__init__(network, **kwargs)

network_plot

Plot a static network.

This function generates a static plot of the network with various output formats including interactive HTML with d3js, tex file with tikz code, PDF from the tex source, and PNG based on matplotlib. The appearance of the plot can be modified using keyword arguments.

Parameters:

Name	Type	Description	Default
network	pathpyG.core.Graph.Graph	A Graph object to be plotted.	required
**kwargs	typing.Any	Keyword arguments to modify the appearance of the plot. Defaults to no attributes. For details see below.	{}

Returns:

Type	Description
pathpyG.visualisations.network_plots.NetworkPlot	A plot object, the type of which depends on the output format chosen.

Keyword Arguments to modify the appearance of the plot

Nodes:

• node_size : diameter of the node

• node_color : The fill color of the node. Possible values are:

  • A single color string referred to by name, RGB or RGBA code, for instance red or #a98d19 or (12,34,102).

  • A sequence of color strings referred to by name, RGB or RGBA code, which will be used for each point's color recursively. For instance ['green', 'yellow'] all points will be filled in green or yellow, alternatively.

  • A column name or position whose values will be used to color the marker points according to a colormap.

• node_cmap : Colormap for node colors. If node colors are given as int or float values the color will be assigned based on a colormap. Per default the color map goes from red to green. Matplotlib colormaps or seaborn color palettes can be used to style the node colors.

• node_opacity : fill opacity of the node. The default is 1. The range of the number lies between 0 and 1. Where 0 represents a fully transparent fill and 1 a solid fill.

Edges

• edge_size : width of the edge

• edge_color : The line color of the edge. Possible values are:

  • A single color string referred to by name, RGB or RGBA code, for instance red or #a98d19 or (12,34,102).

  • A sequence of color strings referred to by name, RGB or RGBA code, which will be used for each point's color recursively. For instance ['green','yellow'] all points will be filled in green or yellow, alternatively.

  • A column name or position whose values will be used to color the marker points according to a colormap.

• edge_cmap : Colormap for edge colors. If node colors are given as int or float values the color will be assigned based on a colormap. Per default the color map goes from red to green. Matplotlib colormaps or seaborn color palettes can be used to style the edge colors.

• edge_opacity : line opacity of the edge. The default is 1. The range of the number lies between 0 and 1. Where 0 represents a fully transparent fill and 1 a solid fill.

General

• keep_aspect_ratio

• margin

• layout

Source code in src/pathpyG/visualisations/network_plots.py

def network_plot(network: Graph, **kwargs: Any) -> NetworkPlot:
    """Plot a static network.

    This function generates a static plot of the network with various output
    formats including interactive HTML with d3js, tex file with tikz code, PDF
    from the tex source, and PNG based on matplotlib. The appearance of the
    plot can be modified using keyword arguments.

    Args:
        network (Graph): A `Graph` object to be plotted.
        **kwargs (Any): Keyword arguments to modify the appearance of the
            plot. Defaults to no attributes. For details see below.

    Returns:
        A plot object, the type of which depends on the output format chosen.


    # Keyword Arguments to modify the appearance of the plot
    **Nodes:**

    - `node_size` : diameter of the node

    - `node_color` : The fill color of the node. Possible values are:

        - A single color string referred to by name, RGB or RGBA code, for
          instance `red` or `#a98d19` or `(12,34,102)`.

        - A sequence of color strings referred to by name, RGB or RGBA code,
          which will be used for each point's color recursively. For
          instance `['green', 'yellow']` all points will be filled in green or
          yellow, alternatively.

        - A column name or position whose values will be used to color the
          marker points according to a colormap.

    - `node_cmap` : Colormap for node colors. If node colors are given as int
      or float values the color will be assigned based on a colormap. Per
      default the color map goes from red to green. Matplotlib colormaps or
      seaborn color palettes can be used to style the node colors.

    - `node_opacity` : fill opacity of the node. The default is 1. The range
      of the number lies between 0 and 1. Where 0 represents a fully
      transparent fill and 1 a solid fill.


    **Edges**

    - `edge_size` : width of the edge

    - `edge_color` : The line color of the edge. Possible values are:

        - A single color string referred to by name, RGB or RGBA code, for
          instance `red` or `#a98d19` or `(12,34,102)`.

        - A sequence of color strings referred to by name, RGB or RGBA
          code, which will be used for each point's color recursively. For
          instance `['green','yellow']` all points will be filled in green or
          yellow, alternatively.

        - A column name or position whose values will be used to color the
          marker points according to a colormap.

    - `edge_cmap` : Colormap for edge colors. If node colors are given as int
      or float values the color will be assigned based on a colormap. Per
      default the color map goes from red to green. Matplotlib colormaps or
      seaborn color palettes can be used to style the edge colors.

    - `edge_opacity` : line opacity of the edge. The default is 1. The range
      of the number lies between 0 and 1. Where 0 represents a fully
      transparent fill and 1 a solid fill.


    **General**

    - `keep_aspect_ratio`

    - `margin`

    - `layout`

    """
    return NetworkPlot(network, **kwargs)

static_plot

Plot a static network.

Source code in src/pathpyG/visualisations/network_plots.py

def static_plot(network: Graph, **kwargs: Any) -> NetworkPlot:
    """Plot a static network."""
    return StaticNetworkPlot(network, **kwargs)

temporal_plot

Plot a temporal network.

Temporal properties:

• start : start time of the simulation

• end : end time of the simulation

• delta : time needed for progressing one time step

• intervals : number of numeric intervals

Source code in src/pathpyG/visualisations/network_plots.py

def temporal_plot(network: TemporalGraph, **kwargs: Any) -> NetworkPlot:
    """Plot a temporal network.

    **Temporal properties:**

    - ``start`` : start time of the simulation

    - ``end`` : end time of the simulation

    - ``delta`` : time needed for progressing one time step

    - ``intervals`` : number of numeric intervals

    """
    return TemporalNetworkPlot(network, **kwargs)