Grokking algorithms(第七章)Dijkstra’s algorithm

上一章的广度优先，可以找到最短的路径，本章的算法，可以找到最快的路径，如下图

 Working with Dijkstra’s algorithm

 Terminology

graph中每一条edge都有一个weight,那么叫weighted graph,没有edge的话，就叫unweighted graph.

在unweighted graph中找出最短路径，使用breadth-first搜索。

在weighted graph中找出最短路径，使用Dijkstra 算法

Negative-weight edges

又edge为负的情况，如下图，lp到poster为-7

不能用Dijkstra’s 算法，可以使用Bellman-Ford algorithm

Implementation

下图为算法作用的图

cost of the node的定义：The cost is how long it takes to get to that node from the start.(从start点到node点的距离)

假如需要更新node B邻域(B指向Fin，fin为B的邻域，因为fin没有箭头指出去，所以fin没有邻域)：以node A举例：Here, you’re calculating how long it would take to get to node A if you went Start > node B > node A, instead of Start > node A.(是计算从start->B->A的cost)

第一步 定义graph

graph = {}
graph[“start”] = {}
graph[“start”][“a”] = 6
graph[“start”][“b”] = 2
graph[“a”] = {}
graph[“a”][“fn”] = 1
graph[“b”] = {}
graph[“b”][“a”] = 3
graph[“b”][“fn”] = 5
graph[“fn”] = {} ##The fnish node doesn’t have any neighbors.

 

 

 

可视化上面的代码，如下图

 

 第二部，定义cost

infnity = float(“inf”)
infnity = float(“inf”)
costs = {}
costs[“a”] = 6
costs[“b”] = 2
costs[“fn”] = infnity

可视化如下图所示：

第三部，定义parents:

parents = {}
parents[“a”] = “start”
parents[“b”] = “start”
parents[“fn”] = None

 可视化如下图

第四部，定义一个列表，里面存储已经处理过的node

processed = []

 下面是整个算法的定义

def fnd_lowest_cost_node(costs):
    lowest_cost = float(“inf”)
    lowest_cost_node = None
    for node in costs: Go through each node.
        cost = costs[node]
        if cost < lowest_cost and node not in processed:
        lowest_cost = cost … set it as the new lowest-cost node.
        lowest_cost_node = node
        return lowest_cost_node

这里需要注意cost这里是一个字典，如下代码相等，print的都是a,b,fin

for node in costs:
    print (node)
for node in costs.keys():
    print (node)

 

node = fnd_lowest_cost_node(costs)
while node is not None:
    cost = costs[node]
    neighbors = graph[node]
    for n in neighbors.keys():
        new_cost = cost + neighbors[n]
        if costs[n] > new_cost:
            costs[n] = new_cost
            parents[n] = node
    processed.append(node)
    node = fnd_lowest_cost_node(costs)

编程实现例子

#define graph
graph = {}
graph['start'] = {}
graph['start']['a'] = 5
graph['start']['b'] = 2

graph['a'] = {}
graph['a']['c'] = 4
graph['a']['d'] = 2

graph['b'] = {}
graph['b']['a'] = 8
graph['b']['d'] = 7

graph['c'] = {}
graph['c']['finish'] = 3
graph['c']['d'] = 6

graph['d'] = {}
graph['d']['finish'] = 1

graph['finish'] = {}

#define costs
infnity = float('inf')
costs = {}
costs['a'] = 5
costs['b'] = 2
costs['c'] = infnity
costs['d'] = infnity
costs['finish'] = infnity

#define parents
parents = {}
parents['a'] = 'start'
parents['b'] = 'start'
parents['finish'] = None

#define processed
processed = []


def find_lowest_cost_node(costs):
    lowestCost = infnity
    lowestNode = None
    for node in costs.keys():
        if costs[node]<lowestCost and node not in processed:
            lowestCost = costs[node]
            lowestNode = node
    return lowestNode

node = find_lowest_cost_node(costs)
while node :
    cost = costs[node]
    neighbors = graph[node]
    for neighbor in neighbors.keys():
        currentCost = cost+neighbors[neighbor]
        if currentCost<costs[neighbor]:
            costs[neighbor] = currentCost
            parents[neighbor] = node
    processed.append(node)
    node = find_lowest_cost_node(costs)

print ('cost of finish is {}'.format(costs['finish']))

 

 

由上图中可以看出，finish的cost为8，并且finish的parent为d ,然后d的parent为a，a的parent为start，因此可以得出结论

start->a->d->finish,中间所得出的cost正好为8