Source code for metrics.edges

import pandas as pd
import networkx as nx




class EdgeMetrics:
    """Handles edge-level metric calculations."""



    @staticmethod
    def extract_edge_metrics(dpg_model, nodes_list):
        """
        Extracts metrics from the edges of a DPG model, including:
        - Edge Load Centrality
        - Trophic Differences
        
        Args:
            dpg_model: A NetworkX graph representing the DPG.
            nodes_list: List of (node_id, label) tuples.

        Returns:
            df: A pandas DataFrame containing the metrics for each edge in the DPG.
        """
        # Map node IDs to labels for fast lookup.
        node_id_to_label = {node_id: label for node_id, label in nodes_list}

        # Calculate edge weights (assuming edges have 'weight' attribute).
        edge_weights = nx.get_edge_attributes(dpg_model, "weight")

        # Build edge rows with labels and IDs.
        edge_data_with_labels = []
        for u, v in dpg_model.edges():
            u_label = node_id_to_label.get(u)
            v_label = node_id_to_label.get(v)
            edge_data_with_labels.append([
                f"{u}-{v}",
                edge_weights.get((u, v), 0),
                u_label,
                v_label,
                u,
                v,
            ])

        # Build a DataFrame with edges, labels, and IDs.
        df_edges_with_labels = pd.DataFrame(
            edge_data_with_labels,
            columns=[
                "Edge",
                "Weight",
                "Node_u_label",
                "Node_v_label",
                "Source_id",
                "Target_id",
            ],
        )

        return df_edges_with_labels