text-graph-generalization

commit f6f89de01867e03e24d2a9828af9c3e4d3068518
parent c196be5171521e5c214fd029210e83eba57f5735
Author: Sondre Wold <sondrewold1@gmail.com>
Date:   Thu, 30 May 2024 15:18:26 +0200

Add stats files

Diffstat:
A
calculate_graph_statistics.py
 | 
28
++++++++++++++++++++++++++++

1 file changed, 28 insertions(+), 0 deletions(-)
diff --git a/calculate_graph_statistics.py b/calculate_graph_statistics.py
@@ -0,0 +1,28 @@
+from graph_sampler import preprocess_graph
+import json
+from rdflib.extras.external_graph_libs import rdflib_to_networkx_multidigraph
+from tqdm import tqdm
+import networkx as nx
+import numpy as np
+import pathlib
+import rdflib
+
+
+if __name__ == '__main__':
+    raw_graphs = pathlib.Path("./raw_graphs/")
+    n_nodes = []
+    n_edges = []
+
+    for input_file in tqdm(raw_graphs.iterdir()):
+        rdf_graph = rdflib.Graph()
+        rdf_graph.parse(input_file)
+        original_graph: nx.MultiDiGraph = rdflib_to_networkx_multidigraph(rdf_graph)
+        filtered_graph, possible_relations = preprocess_graph(original_graph)
+        n_nodes.append(len(filtered_graph.nodes()))
+        n_edges.append(len(filtered_graph.edges()))
+    print(f"Mean nodes: {np.mean(n_nodes)}, std: {np.std(n_nodes)}")
+    print(f"Mean edges: {np.mean(n_edges)}, std: {np.std(n_edges)}")
+
+    stats = {"nodes":[np.mean(n_nodes), np.std(n_nodes)], "edges": [np.mean(n_edges), np.std(n_edges)]}
+    with open("./graph_stats.json", "w") as f:
+        json.dump(stats, f)