Merge branch 'feature/graph-algorithms' into develop

resolve conflicts

Author: codejsha

README.md

@@ -260,7 +260,7 @@ algorithm DFS-VISIT(G, u):
     u.finished = time
 ```
 
-- Dijkstra's algorithm: [c++](cpp-algorithm/src/graph), [java](java-algorithm/src/main/java/com/example/algorithm/graph) | A single source the shortest path algorithm that handle non-negative edge weights. It find the shortest path between two vertices in a graph.
+- Dijkstra's algorithm, CCSP#4.5.1: [c++](cpp-algorithm/src/graph),[python](python-algorithm/algorithm/graph/test)(test), [java](java-algorithm/src/main/java/com/example/algorithm/graph) | A single source the shortest path algorithm that handle non-negative edge weights. It find the shortest path between two vertices in a graph.
 
 ```txt
 algorithm Dijkstra(G, source):
@@ -347,6 +347,7 @@ algorithm Prim(G, root):
 **Examples**
 
 - Maze problem: [java](java-algorithm/src/main/java/com/example/algorithm/graph) | A maze problem is that find a path from the start to the goal. The maze is represented by a graph. The start and the goal are represented by vertices. The path is represented by a sequence of vertices.
+- Minimum spanning tree (Kruskal, Prim, Boruvka), CCSP#4.4.2: [python](python-algorithm/algorithm/graph/test)(test) | Find the minimum spanning tree of a graph.
 
 [:arrow_up_small: back to toc](#table-of-contents)
 

python-algorithm/algorithm/graph/test/graph_data_utils.py

@@ -0,0 +1,85 @@
+import networkx
+
+seattle = 'Seattle'
+san_francisco = 'San Francisco'
+los_angeles = 'Los Angeles'
+riverside = 'Riverside'
+phoenix = 'Phoenix'
+dallas = 'Dallas'
+houston = 'Houston'
+atlanta = 'Atlanta'
+miami = 'Miami'
+chicago = 'Chicago'
+detroit = 'Detroit'
+boston = 'Boston'
+new_york = 'New York'
+philadelphia = 'Philadelphia'
+washington = 'Washington'
+
+
+def create_city_graph() -> networkx.Graph:
+    edges = [
+        (seattle, chicago),
+        (seattle, san_francisco),
+        (san_francisco, riverside),
+        (san_francisco, los_angeles),
+        (los_angeles, riverside),
+        (los_angeles, phoenix),
+        (riverside, phoenix),
+        (riverside, chicago),
+        (phoenix, dallas),
+        (phoenix, houston),
+        (dallas, chicago),
+        (dallas, atlanta),
+        (dallas, houston),
+        (houston, atlanta),
+        (houston, miami),
+        (atlanta, chicago),
+        (atlanta, washington),
+        (atlanta, miami),
+        (miami, washington),
+        (chicago, detroit),
+        (detroit, boston),
+        (detroit, washington),
+        (detroit, new_york),
+        (boston, new_york),
+        (new_york, philadelphia),
+        (philadelphia, washington)
+    ]
+    graph = networkx.Graph()
+    graph.add_edges_from(edges)
+    return graph
+
+
+def create_weighted_city_graph() -> networkx.Graph:
+    weighted_edges = [
+        (seattle, chicago, 1737),
+        (seattle, san_francisco, 678),
+        (san_francisco, riverside, 386),
+        (san_francisco, los_angeles, 348),
+        (los_angeles, riverside, 50),
+        (los_angeles, phoenix, 357),
+        (riverside, phoenix, 307),
+        (riverside, chicago, 1704),
+        (phoenix, dallas, 887),
+        (phoenix, houston, 1015),
+        (dallas, chicago, 805),
+        (dallas, atlanta, 721),
+        (dallas, houston, 225),
+        (houston, atlanta, 702),
+        (houston, miami, 968),
+        (atlanta, chicago, 588),
+        (atlanta, washington, 543),
+        (atlanta, miami, 604),
+        (miami, washington, 923),
+        (chicago, detroit, 238),
+        (detroit, boston, 613),
+        (detroit, washington, 396),
+        (detroit, new_york, 482),
+        (boston, new_york, 190),
+        (new_york, philadelphia, 81),
+        (philadelphia, washington, 123)
+    ]
+    graph = networkx.Graph()
+    graph.add_weighted_edges_from(weighted_edges)
+    return graph

python-algorithm/algorithm/graph/test/test_breadth_first_search.py

@@ -1,36 +1,7 @@
-import pytest, networkx
-
-
-def create_city_graph() -> networkx.Graph:
-    graph = networkx.Graph()
-    graph.add_edge("Seattle", "Chicago")
-    graph.add_edge("Seattle", "San Francisco")
-    graph.add_edge("San Francisco", "Riverside")
-    graph.add_edge("San Francisco", "Los Angeles")
-    graph.add_edge("Los Angeles", "Riverside")
-    graph.add_edge("Los Angeles", "Phoenix")
-    graph.add_edge("Riverside", "Phoenix")
-    graph.add_edge("Riverside", "Chicago")
-    graph.add_edge("Phoenix", "Dallas")
-    graph.add_edge("Phoenix", "Houston")
-    graph.add_edge("Dallas", "Chicago")
-    graph.add_edge("Dallas", "Atlanta")
-    graph.add_edge("Dallas", "Houston")
-    graph.add_edge("Houston", "Atlanta")
-    graph.add_edge("Houston", "Miami")
-    graph.add_edge("Atlanta", "Chicago")
-    graph.add_edge("Atlanta", "Washington")
-    graph.add_edge("Atlanta", "Miami")
-    graph.add_edge("Miami", "Washington")
-    graph.add_edge("Chicago", "Detroit")
-    graph.add_edge("Detroit", "Boston")
-    graph.add_edge("Detroit", "Washington")
-    graph.add_edge("Detroit", "New York")
-    graph.add_edge("Boston", "New York")
-    graph.add_edge("New York", "Philadelphia")
-    graph.add_edge("Philadelphia", "Washington")
-    return graph
+import networkx
+import pytest
 
+from algorithm.graph.test.graph_data_utils import create_city_graph
 
 city_graph = create_city_graph()
 
@@ -39,21 +10,21 @@ def create_city_graph() -> networkx.Graph:
 @pytest.mark.parametrize(
     argnames='graph, source, expected_neighbors',
     argvalues=[
-        (city_graph, "Seattle", ["Chicago", "San Francisco"]),
-        (city_graph, "San Francisco", ["Seattle", "Riverside", "Los Angeles"]),
-        (city_graph, "Los Angeles", ["San Francisco", "Riverside", "Phoenix"]),
-        (city_graph, "Riverside", ["San Francisco", "Los Angeles", "Phoenix", "Chicago"]),
-        (city_graph, "Phoenix", ["Los Angeles", "Riverside", "Dallas", "Houston"]),
-        (city_graph, "Chicago", ["Seattle", "Riverside", "Dallas", "Atlanta", "Detroit"]),
-        (city_graph, "Boston", ["Detroit", "New York"]),
-        (city_graph, "New York", ["Detroit", "Boston", "Philadelphia"]),
-        (city_graph, "Atlanta", ["Dallas", "Houston", "Chicago", "Washington", "Miami"]),
-        (city_graph, "Miami", ["Houston", "Atlanta", "Washington"]),
-        (city_graph, "Dallas", ["Phoenix", "Chicago", "Atlanta", "Houston"]),
-        (city_graph, "Houston", ["Phoenix", "Dallas", "Atlanta", "Miami"]),
-        (city_graph, "Detroit", ["Chicago", "Boston", "Washington", "New York"]),
-        (city_graph, "Philadelphia", ["New York", "Washington"]),
-        (city_graph, "Washington", ["Atlanta", "Miami", "Detroit", "Philadelphia"])
+        (city_graph, 'Seattle', ['Chicago', 'San Francisco']),
+        (city_graph, 'San Francisco', ['Seattle', 'Riverside', 'Los Angeles']),
+        (city_graph, 'Los Angeles', ['San Francisco', 'Riverside', 'Phoenix']),
+        (city_graph, 'Riverside', ['San Francisco', 'Los Angeles', 'Phoenix', 'Chicago']),
+        (city_graph, 'Phoenix', ['Los Angeles', 'Riverside', 'Dallas', 'Houston']),
+        (city_graph, 'Chicago', ['Seattle', 'Riverside', 'Dallas', 'Atlanta', 'Detroit']),
+        (city_graph, 'Boston', ['Detroit', 'New York']),
+        (city_graph, 'New York', ['Detroit', 'Boston', 'Philadelphia']),
+        (city_graph, 'Atlanta', ['Dallas', 'Houston', 'Chicago', 'Washington', 'Miami']),
+        (city_graph, 'Miami', ['Houston', 'Atlanta', 'Washington']),
+        (city_graph, 'Dallas', ['Phoenix', 'Chicago', 'Atlanta', 'Houston']),
+        (city_graph, 'Houston', ['Phoenix', 'Dallas', 'Atlanta', 'Miami']),
+        (city_graph, 'Detroit', ['Chicago', 'Boston', 'Washington', 'New York']),
+        (city_graph, 'Philadelphia', ['New York', 'Washington']),
+        (city_graph, 'Washington', ['Atlanta', 'Miami', 'Detroit', 'Philadelphia'])
     ])
 def test_graph_neighbors(benchmark, graph, source, expected_neighbors):
     """
@@ -71,9 +42,9 @@ def test_graph_neighbors(benchmark, graph, source, expected_neighbors):
 @pytest.mark.parametrize(
     argnames='graph, source, distance, expected_nodes',
     argvalues=[
-        (city_graph, "Boston", 1, ["Detroit", "New York"]),
-        (city_graph, "Boston", 2, ["Chicago", "Washington", "Philadelphia"]),
-        (city_graph, "Boston", 3, ["Seattle", "Riverside", "Dallas", "Atlanta", "Miami"]),
+        (city_graph, 'Boston', 1, ['Detroit', 'New York']),
+        (city_graph, 'Boston', 2, ['Chicago', 'Washington', 'Philadelphia']),
+        (city_graph, 'Boston', 3, ['Seattle', 'Riverside', 'Dallas', 'Atlanta', 'Miami']),
     ],
     ids=['distance1', 'distance2', 'distance3'])
 def test_graph_breadth_first_search(benchmark, graph, source, distance, expected_nodes):

python-algorithm/algorithm/graph/test/test_dijkstra.py

@@ -0,0 +1,46 @@
+import networkx
+import pytest
+
+from algorithm.graph.test.graph_data_utils import create_weighted_city_graph
+
+city_graph = create_weighted_city_graph()
+
+
+@pytest.mark.benchmark(group='graph_dijkstra')
+@pytest.mark.parametrize(
+    argnames='graph, source, target, expected_path',
+    argvalues=[
+        (city_graph, 'Los Angeles', 'Boston', ['Los Angeles', 'Riverside', 'Chicago', 'Detroit', 'Boston']),
+    ],
+    ids=['los_angeles_to_boston'])
+def test_graph_dijkstra_path(benchmark, graph, source, target, expected_path):
+    """
+    Find the shortest path between two nodes in a graph using Dijkstra's algorithm.
+    :param benchmark: benchmark fixture
+    :param graph: city graph of the United States
+    :param source: source city node
+    :param target: target city node
+    :param expected_path: expected shortest path
+    """
+    shortest_path = benchmark(networkx.dijkstra_path, graph, source, target)
+    assert shortest_path == expected_path
+
+
+@pytest.mark.benchmark(group='graph_dijkstra')
+@pytest.mark.parametrize(
+    argnames='graph, source, target, expected_distance',
+    argvalues=[
+        (city_graph, 'Los Angeles', 'Boston', 2605),
+    ],
+    ids=['los_angeles_to_boston'])
+def test_graph_dijkstra_distance(benchmark, graph, source, target, expected_distance):
+    """
+    Find the shortest distance between two nodes in a graph using Dijkstra's algorithm.
+    :param benchmark: benchmark fixture
+    :param graph: city graph of the United States
+    :param source: source city node
+    :param target: target city node
+    :param expected_distance: expected shortest distance (weight)
+    """
+    total_weight = benchmark(networkx.dijkstra_path_length, graph, source, target)
+    assert total_weight == expected_distance

python-algorithm/algorithm/graph/test/test_minimum_spanning_tree.py

@@ -0,0 +1,44 @@
+import matplotlib.pyplot as plt
+import networkx
+import pytest
+
+from algorithm.graph.test.graph_data_utils import create_weighted_city_graph
+
+city_graph = create_weighted_city_graph()
+
+
+@pytest.mark.skip(reason='This test is for visualization only')
+def test_graph_mst_temp():
+    mst = networkx.minimum_spanning_tree(city_graph, algorithm='kruskal')
+    pos = networkx.spring_layout(mst)
+    networkx.draw_networkx_nodes(mst, pos)
+    networkx.draw_networkx_edges(mst, pos, width=1)
+    networkx.draw_networkx_labels(mst, pos, font_size=10)
+    plt.show()
+
+
+city_edges = {('Seattle', 'San Francisco'), ('San Francisco', 'Los Angeles'), ('Los Angeles', 'Riverside'),
+              ('Riverside', 'Phoenix'), ('Phoenix', 'Dallas'), ('Dallas', 'Houston'), ('Houston', 'Atlanta'),
+              ('Atlanta', 'Miami'), ('Atlanta', 'Washington'), ('Washington', 'Philadelphia'),
+              ('Philadelphia', 'New York'), ('New York', 'Boston'), ('Washington', 'Detroit'), ('Detroit', 'Chicago')}
+
+
+@pytest.mark.benchmark(group='graph_minimum_spanning_tree')
+@pytest.mark.parametrize(
+    argnames='graph, algorithm, expected_total_weight, expected_edges',
+    argvalues=[
+        (city_graph, 'kruskal', 5372, city_edges),
+        (city_graph, 'prim', 5372, city_edges),
+        (city_graph, 'boruvka', 5372, city_edges),
+    ],
+    ids=['kruskal', 'prim', 'boruvka'])
+def test_graph_mst(benchmark, graph, algorithm, expected_total_weight, expected_edges):
+    mst = benchmark(networkx.minimum_spanning_tree, graph, algorithm=algorithm)
+
+    mst_total_weight = sum(d['weight'] for u, v, d in mst.edges(data=True))
+    assert expected_total_weight == mst_total_weight
+
+    mst_edges = set((u, v) for u, v, d in mst.edges(data=True))
+    normalized_set1 = {tuple(sorted(edge)) for edge in expected_edges}
+    normalized_set2 = {tuple(sorted(edge)) for edge in mst_edges}
+    assert normalized_set1 == normalized_set2

python-algorithm/pyproject.toml

@@ -11,6 +11,7 @@ python = ">=3.11,<3.13"
 numpy = "^1.26.0"
 scipy = "^1.12.0"
 networkx = "^3.2.1"
+matplotlib = "^3.8.2"
 
 [tool.poetry.group.test.dependencies]
 pytest = "^8.0.0"