money-mining-python/cal.py

# -*- coding:utf-8 -*-

import config
import re
import numpy as np

def merge_stem(a_stem:list, b_stem:list):
    """
    合并词根
    """
    return list(set(a_stem).union(set(b_stem)))

def gen_word_vec(a_word:str, b_word:str, stem:list):
    """
    生成词向量
    """
    a_vec, b_vec = [], []
    for word in stem:
        # if re.findall(word, config.RE_SPECIAL_SIMBOL):
        if word in config.RE_SPECIAL_SIMBOL:
            word = "\\" + word
        if word == "c++":
            word = "c\\+\\+"
        a_vec.append(len(re.findall(word, a_word)))
        b_vec.append(len(re.findall(word, b_word)))
    return a_vec, b_vec

def col_sim(vec1, vec2):
    """
    计算余弦相似性
    """
    return vec1.dot(vec2) / (np.linalg.norm(vec1) * np.linalg.norm(vec2))


def cal_cos_sim(a_word:str, a_stem:list, b_word:str, b_stem:list):
    """
    计算余弦相似性
    """
    union_stem = merge_stem(a_stem, b_stem)
    a_vec, b_vec = gen_word_vec(a_word, b_word, union_stem)
    val = col_sim(np.array(a_vec), np.array(b_vec))
    return val


if __name__ == "__main__":

    # a_word= "QQ邮箱格式怎么写"
    # b_word= "QQ邮箱格式如何写"
    # a_word_root = ['QQ', '邮箱', '格式', '怎么', '写']
    # b_word_root = ['QQ', '邮箱', '格式', '如何', '写']
    # print(cal_cos_sim(a_word, a_word_root, b_word, b_word_root))

     # 合并词根，用于生成词向量
    # union_word_root = merge_stem(a_word_root, b_word_root)
    # print(union_word_root)

    # # 生成词向量
    # a_vec, b_vec = gen_word_vec(a_word, b_word, union_word_root)
    # print(a_vec)
    # print(b_vec)
    # # a_vec = [1,1,1,1,0,1]
    # # b_vec = [1,1,1,0,1,1]
    # print(col_sim(np.array(a_vec), np.array(b_vec)))

    # s = "0,腋下长了一个小疙瘩是什么东西,['腋下', '长', '了', '一个', '小', '疙瘩', '是', '什么', '东西']"
    # s_r = r"'([^,]*)'"
    # pattern = re.compile(s_r, re.I)
    # for i in pattern.findall(s):
    #     print(i)
    
    # s_r = r"([\d]*),(.*),\["
    # pattern = re.compile(s_r, re.I)
    # m = pattern.match(s)
    # for i in m.groups():
    #     print(i)

    # import mmap
    # with open(config.KEY_FILE, "r", encoding=config.ENCODING_CHARSET) as f_key, \
    #     mmap.mmap(f_key.fileno(), 0, access=mmap.ACCESS_READ) as f_key_mmap:

    #     key_info_re = r"([\d]*),(.*),\["
    #     key_info_pattern = re.compile(key_info_re, re.I)

    #     s_r = r"'([^,]*)'"
    #     s_pattern = re.compile(s_r, re.I)

    #     a_line = f_key_mmap.readline().decode("UTF-8")
    #     b_line = f_key_mmap.readline().decode("UTF-8")

    #     a_m = key_info_pattern.match(a_line)
    #     a_key = a_m.group(2)
    #     a_stem = s_pattern.findall(a_line)
    #     print(a_stem)

    #     b_m = key_info_pattern.match(b_line)
    #     b_key = b_m.group(2)
    #     b_stem = s_pattern.findall(b_line)
    #     print(b_stem)

    #     print(cal_cos_sim(a_key, a_stem, b_key, b_stem))

    # a_key = "吃什么东西减肥最快"
    # a_stem = ['吃', '什么', '东西', '减肥', '最快']

    # b_key="vc++读写什么文件最快"
    # b_stem =['v', 'c++', '读写', '什么', '文件', '最快']
    # print(cal_cos_sim(a_key, a_stem, b_key, b_stem))

    # print(re.findall("c\\+\\+", "vc++读写什么文件最快"))
    
    
    print("".join([".", "?", "^", "$", "*", "+", "\\", "[", "]", "|", "{", "}", "(", ")"]))
    # s = r"([.?^$*+\[]|{}()])"
    s=r"([\\])"
    re.findall(s, "vc++读写什么文件最快")
    print(re.findall(s, "vc++读写什么文件\最快"))