a1a430fa42fa58cb3ba9d93f44749319928b78c8,dirty_cat/string_distances.py,,ngram_similarity,#,90

Before Change

     n-gram similarity between two strings
    
    ngrams1 = get_ngrams(string1, n)
    count_dict1 = {}
    for ngram in ngrams1:
        count_dict1.setdefault(ngram, 0)
        count_dict1[ngram] += 1

After Change

    ngrams2 = get_ngrams(string2, n)
    count2 = Counter(ngrams2)

    samegrams = sum((count1 & count2).values())
    allgrams = len(ngrams1) + len(ngrams2)
    similarity = samegrams/(allgrams - samegrams)
    return similarity

Instances