probinet.visualization.plot

probinet.visualization.plot#

It provides a set of plotting functions for visualizing the results of the generative models.

Functions

extract_bridge_properties(i, color, U[, ...])

Extract the properties of the bridges of a node i.

mapping(G, A)

Map the nodes of a graph G to the nodes of a graph A.

plot_A(A[, cmap])

Plot the adjacency tensor produced by the generative algorithm.

plot_L(values[, indices, k_i, xlab, ylabel, ...])

Plot the log-likelihood.

plot_M(M[, cmap, figsize, fontsize])

Plot the M matrix produced by the generative algorithm.

plot_adjacency(Bd, M_marginal, ...[, cm])

Plot the adjacency matrix and its reconstruction given by the marginal and the conditional expected values.

plot_adjacency_samples(Bdata, Bsampled[, cm])

Plot the adjacency matrix and five sampled networks.

plot_graph(graph, M_marginal, M_conditional, ...)

Plot a graph and its reconstruction given by the marginal and the conditional expected values.

plot_hard_membership(graph, communities, ...)

Plot a graph with nodes colored by their hard memberships.

plot_precision_recall(conf_matrix[, cm])

Plot precision and recall of a given confusion matrix.

plot_soft_membership(graph, thetas, pos, ...)

Plot a graph with nodes colored by their mixed (soft) memberships.
probinet.visualization.plot.extract_bridge_properties(i: int, color: dict, U: ndarray, threshold: float = 0.2) → Tuple[ndarray, list][source]#

Extract the properties of the bridges of a node i.

Parameters:

  • i (int) – Index of the node.

  • color (dict) – Dictionary with the colors of the nodes.

  • U (ndarray) – Out-going membership matrix.

  • threshold (float) – Threshold for the membership values.

Returns:

  • wedge_sizes (ndarray) – Sizes of the wedges.

  • wedge_colors (list) – Colors of the wedges.

probinet.visualization.plot.mapping(G: DiGraph, A: DiGraph) → DiGraph[source]#

Map the nodes of a graph G to the nodes of a graph A.

Parameters:

  • G (nx.DiGraph) – Graph to be mapped.

  • A (nx.DiGraph) – Graph to be mapped to.

Returns:

G – Graph G with the nodes mapped to the nodes of A.

Return type:

nx.DiGraph

probinet.visualization.plot.plot_A(A: List, cmap: str = 'Blues') → List[Figure][source]#

Plot the adjacency tensor produced by the generative algorithm.

Parameters:

  • A (list) – List of scipy sparse matrices, one for each layer.

  • cmap (Matplotlib object) – Colormap used for the plot.

Returns:

figures – List of matplotlib figure objects.

Return type:

list

probinet.visualization.plot.plot_L(values: List, indices: List | None = None, k_i: int = 0, xlab: str = 'Iterations', ylabel: str = 'Log-likelihood values', figsize: tuple = (10, 5), int_ticks: bool = False) → Figure[source]#

Plot the log-likelihood. :param values: List of log-likelihood values. :type values: list :param indices: List of indices. :type indices: list :param k_i: Number of initial iterations to be ignored. :type k_i: int :param xlab: Label of the x-axis. :type xlab: str :param ylabel: :type ylabel: str :param figsize: Figure size. :type figsize: tuple :param int_ticks: Flag to use integer ticks. :type int_ticks: bool

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure

probinet.visualization.plot.plot_M(M: ndarray, cmap: str = 'PuBuGn', figsize: Tuple[int, int] = (7, 7), fontsize: int = 15) → None[source]#

Plot the M matrix produced by the generative algorithm. Each entry is the Poisson mean associated with each pair of nodes in the graph.

Parameters:

  • M (np.ndarray) – NxN M matrix associated with the graph. Contains all the means used for generating edges.

  • cmap (str, optional) – Colormap used for the plot.

  • figsize (Tuple[int, int], optional) – Size of the figure to be plotted.

  • fontsize (int, optional) – Font size to be used in the plot title.

probinet.visualization.plot.plot_adjacency(Bd: ndarray, M_marginal: ndarray, M_conditional: ndarray, nodes: List, cm: str = 'Blues') → Figure[source]#

Plot the adjacency matrix and its reconstruction given by the marginal and the conditional expected values.

Parameters:

  • Bd (ndarray) – Adjacency matrix.

  • M_marginal (ndarray) – Marginal expected values.

  • M_conditional (ndarray) – Conditional expected values.

  • nodes (list) – List of nodes.

  • cm (Matplotlib object) – Colormap used for the plot.

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure

probinet.visualization.plot.plot_adjacency_samples(Bdata: List, Bsampled: List, cm: str = 'Blues') → Figure[source]#

Plot the adjacency matrix and five sampled networks.

Parameters:

  • Bdata (list) – List of adjacency matrices for the data.

  • Bsampled (list) – List of adjacency matrices for sampled networks.

  • cm (Matplotlib object) – Colormap used for the plot.

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure

probinet.visualization.plot.plot_graph(graph: DiGraph, M_marginal: ndarray, M_conditional: ndarray, pos: Dict, node_size: int, node_color: str, edge_color: str, threshold: float = 0.2) → Figure[source]#

Plot a graph and its reconstruction given by the marginal and the conditional expected values.

Parameters:

  • graph (nx.DiGraph) – Graph to be plotted.

  • M_marginal (ndarray) – Marginal expected values.

  • M_conditional (ndarray) – Conditional expected values.

  • pos (dict) – Dictionary with the positions of the nodes.

  • node_size (int) – Size of the nodes.

  • node_color (str) – Color of the nodes.

  • edge_color (str) – Color of the edges.

  • threshold (float) – Threshold for the membership values.

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure

probinet.visualization.plot.plot_hard_membership(graph: DiGraph, communities: Dict, pos: Dict, node_size: ndarray, colors: Dict, edge_color: str) → Figure[source]#

Plot a graph with nodes colored by their hard memberships.

Parameters:

  • graph (nx.DiGraph) – Graph to be plotted.

  • communities (Dict) – Dictionary with the communities.

  • pos (Dict) – Dictionary with the positions of the nodes.

  • node_size (ndarray) – Array with the sizes of the nodes.

  • colors (Dict) – Dictionary with the colors of the nodes.

  • edge_color (str) – Color of the edges.

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure

probinet.visualization.plot.plot_precision_recall(conf_matrix: ndarray, cm: str = 'Blues') → Figure[source]#

Plot precision and recall of a given confusion matrix.

Parameters:

  • conf_matrix (ndarray) – Confusion matrix.

  • cm (Matplotlib object) – Colormap used for the plot.

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure

probinet.visualization.plot.plot_soft_membership(graph: DiGraph, thetas: Dict, pos: Dict, node_size: ndarray, colors: Dict, edge_color: str) → Figure[source]#

Plot a graph with nodes colored by their mixed (soft) memberships.

Parameters:

  • graph (nx.DiGraph) – Graph to be plotted.

  • thetas (Dict) – Dictionary with the mixed memberships.

  • pos (Dict) – Dictionary with the positions of the nodes.

  • node_size (ndarray) – Array with the sizes of the nodes.

  • colors (Dict) – Dictionary with the colors of the nodes.

  • edge_color (str) – Color of the edges.

Returns:

fig – The matplotlib figure object.

Return type:

plt.Figure