def mst_kruskal(graph):
# Kruskal's algorithm explanation https://www.youtube.com/watch?v=71UQH7Pr9kU
mst = Graph(False)
min_edge_heap = [
] # use as priority queue to select and edge with minimum weight
# 1. Sort the edges in ascending order of weights
for edge in graph.get_edges():
heapq.heappush(min_edge_heap, edge)
input_graph_num_vertices = len(graph.get_vertices())
# 2. Keep adding edges until MST will reach all vertices
while (len(min_edge_heap) > 0) and (len(mst.get_vertices()) <
input_graph_num_vertices):
# 3. Select an edge with minimum weight (greedy algorithm)
min_edge = heapq.heappop(min_edge_heap)
start_vertex = min_edge.get_start_vertex()
end_vertex = min_edge.get_end_vertex()
# 4. Add edge in MST and check if it form a cycle then remove edge, otherwise leave it in MST
mst.add_vertex(start_vertex.get_label())
mst.add_vertex(end_vertex.get_label())
start_label = start_vertex.get_label()
end_label = end_vertex.get_label()
weight = min_edge.get_weight()
mst.add_edge(start_label, end_label, weight)
if has_cycle(mst):
mst.remove_edge(start_label, end_label, weight)
return mst
def mst_prim(graph):
# Prim's algorithm explanation https://www.youtube.com/watch?v=cplfcGZmX7I
# This algorithm return list of edges, so if we have edges we may recreate MST
mst = Graph(False)
min_edge_heap = []
visited_vertices = set()
# 1. Choose arbitary vertex from which MST algorithm start looking for edges
arbitary_vertex = next(iter(graph.get_vertices()))
visited_vertices.add(arbitary_vertex)
mst.add_vertex(arbitary_vertex)
# 2. Find all adjacent edges of arbitary vertex and add them to heap
for edge in graph.get_vertex(arbitary_vertex).get_outbound_edges():
heapq.heappush(min_edge_heap, edge)
input_graph_num_vertices = len(graph.get_vertices())
while len(mst.get_vertices()) < input_graph_num_vertices:
# 3. Select an edge with minimum weight (greedy algorithm)
while True:
min_edge = heapq.heappop(min_edge_heap)
min_vertex = min_edge.get_end_vertex()
min_vertex_label = min_vertex.get_label()
if min_vertex_label not in visited_vertices:
break
# 4. Mark selected vertex as visited and add selected edge to MST
visited_vertices.add(min_vertex_label)
mst.add_vertex(min_vertex_label)
mst.add_edge(min_edge.get_start_vertex().get_label(),
min_edge.get_end_vertex().get_label(), min_edge.get_weight())
# 4. Find all adjacent edges of min vertex and add them to heap
for edge in min_vertex.get_outbound_edges():
heapq.heappush(min_edge_heap, edge)
return mst
for edge in current_vertex.get_outbound_edges():
adjacent_vertex = edge.get_end_vertex()
if adjacent_vertex not in visited_vertices:
adjacent_vertices.append(adjacent_vertex)
# if necessary we may do some manipulation with adjacent_vertices, e.g. sort them
# 5. add all adjacent vertices to the queue(BFS)
queue.extend(adjacent_vertices)
return result
if __name__ == "__main__":
graph = Graph()
graph.add_vertex("Jhon")
graph.add_vertex("Sophia")
graph.add_vertex("Emma")
graph.add_vertex("Mark")
graph.add_vertex("Alice")
graph.add_vertex("Jeff")
graph.add_vertex("George")
graph.add_edge("Jhon", "Sophia")
graph.add_edge("Jhon", "Emma")
graph.add_edge("Jhon", "Mark")
graph.add_edge("Sophia", "Emma")
graph.add_edge("Sophia", "Alice")
graph.add_edge("Emma", "Sophia")
graph.add_edge("Emma", "Jeff")
graph.add_edge("Jeff", "George")
min_vertex = graph.get_vertex(min_vertex_label)
for edge in min_vertex.get_outbound_edges():
new_distance = min_distance + edge.get_weight()
# 4. write the new distance to adjacent vertex if we have found a better path
adjacent_vertex = edge.get_end_vertex()
if new_distance < distances[adjacent_vertex]:
distances[adjacent_vertex] = new_distance
return distances
if __name__ == "__main__":
graph = Graph()
graph.add_vertex("a")
graph.add_vertex("b")
graph.add_vertex("c")
graph.add_vertex("d")
graph.add_vertex("e")
graph.add_vertex("f")
graph.add_edge("a", "b", 2)
graph.add_edge("a", "c", 4)
graph.add_edge("b", "c", 1)
graph.add_edge("c", "d", 5)
graph.add_edge("c", "e", 3)
graph.add_edge("d", "e", 1)
graph.add_edge("e", "f", 8)
display_graph(graph, "Input graph for Dijkstra's algorithm")
adjacent_vertex_label] - 1
if indegree_dict[adjacent_vertex_label] == 0:
zero_indegree_vertices.put(adjacent_vertex_label)
if len(topological_order) != len(graph.get_vertices()):
raise ValueError("This graph is cyclic")
return topological_order
if __name__ == "__main__":
dag = Graph() # directed acyclic graph
# vertices
dag.add_vertex("0")
dag.add_vertex("1")
dag.add_vertex("2")
dag.add_vertex("3")
dag.add_vertex("4")
dag.add_vertex("5")
dag.add_vertex("6")
dag.add_vertex("7")
dag.add_vertex("8")
# edges
dag.add_edge("0", "1")
dag.add_edge("1", "2")
dag.add_edge("1", "5")
dag.add_edge("2", "3")
dag.add_edge("2", "4")