Question

我的编程任务是，给定一个输入文件words.txt，创建一个图形，其中节点是单词本身，边缘是一个字母不同的单词。（例如，＆＃34; abcdef＆＃34;和＆＃34; abcdeg＆＃34;会有优势。）

对于问题的后期阶段，我确实需要使用邻接列表实现（因此没有邻接矩阵）。如果可能，我不想使用外部库。

我的代码功能齐全，但运行大约需要7秒钟（约5000字）。我需要帮助进一步优化它。我做的主要优化是对字符串中的每个字符使用{＆＃34; index，char＆＃34;}的集合并比较集合，这比sum(x!=y for x,y in zip(word1, word2))

更快。

这是我到目前为止的代码：

def createGraph(filename):
    """Time taken to generate graph: {} seconds."""
    wordReference = {} # Words and their smartsets (for comparisons)
    wordEdges = {} # Words and their edges (Graph representation)
    edgeCount = 0 # Count of edges
    with open(filename) as f:
        for raw in f.readlines(): # For each word in the file (Stripping trailing \n)
            word = raw.strip()
            wordReference[word] = set("{}{}".format(letter, str(i)) for i, letter in enumerate(word)) # Create smartsets in wordReference
            wordEdges[word] = set() # Create a set to contain the edges
            for parsed in wordEdges: # For each of the words we've already parsed
                if len(wordReference[word] & wordReference[parsed]) == 5: # If compare smartSets to see if there should be an edge
                    wordEdges[parsed].add(word) #Add edge parsed -> word
                    wordEdges[word].add(parsed) #Add edge word -> parsed
                    edgeCount += 1 #Add 1 to edgesCount
    return wordEdges, edgeCount, len(wordReference) #Return dictionary of words and their edges

提示优化我的python图形创建代码

0 个答案: