stock.csv
    movie_data_cleaned.csv

    1. '''
    2. file: pandas_def.py
    3. '''
    4. import pandas as pd
    5. import numpy as np
    7. pd.set_option('display.max_columns', None)
    8. pd.set_option('display.max_rows', None)
    9. file = "movie_data_cleaned.csv"
    12. def movie_year_amount_tj():
    13.     data = pd.read_csv(file)
    14.     data['release_date'] = pd.to_datetime(data['release_date'])
    15.     data = data.set_index(data['release_date'])
    16.     data_year_tj = data['release_date'].resample('Y').count()
    17.     return pd.DataFrame(data_year_tj)
    20. def country_year_tj():
    21.     data = pd.read_csv(file,
    22.                        usecols=['title', 'country', 'language', 'release_date', 'average'])
    23.     data = data[['title', 'country', 'language', 'release_date', 'average']]
    24.     # 各国每年的电影产量
    25.     data['country'] = data['country'].str.strip(' ')
    26.     data['country'] = data['country'].fillna(value='')
    27.     country_list = []
    28.     for c in data['country']:
    29.         c_list = c.split(' / ')
    30.         for label in c_list:  # 123,2,3 -> 1,2,3
    31.             country_list.append(label)
    32.     country_list = list(set(country_list))
    33.     country_list.remove('')
    34.     country_list.remove('美国/澳大利亚')
    35.     country_list.remove('捷克斯洛伐克/捷克')
    36.     country_list.remove('中国大陆')  # 只统计中国
    37.     country_list.remove('中国香港')
    38.     country_list.remove('中国台湾')
    39.     data['release_date'] = pd.to_datetime(data['release_date'])
    40.     data = data.set_index(data['release_date'])
    41.     count = 0
    42.     tj = pd.DataFrame(data['release_date'].resample('Y').count())
    43.     tj = tj.drop(columns='release_date')
    44.     for label in country_list:
    45.         temp = data[data['country'].str.contains(label)]
    46.         print("=====================================")
    47.         print("标签=", label)
    48.         print("总频数=", len(temp))
    49.         count += len(temp)
    50.         tj[label] = temp['release_date'].resample('Y').count()
    51.     tj = tj.fillna(value=0)
    52.     return tj
    55. def language_tj():
    56.     data = pd.read_csv(file,
    57.                        usecols=['title', 'country', 'language', 'release_date', 'average'])
    58.     data = data[['title', 'country', 'language', 'release_date', 'average']]
    59.     # label统计 -> list
    60.     data['language'] = data['language'].str.strip(' ')
    61.     data['language'] = data['language'].fillna(value='')
    62.     lang_list = []
    63.     for l in data['language']:
    64.         l_list = l.split(' / ')
    65.         for label in l_list:  # 123,2,3 -> 1,2,3
    66.             lang_list.append(label)
    67.     lang_list = list(set(lang_list))
    68.     lang_list.remove('')
    69.     lang_list.remove('汉语普通话')
    70.     # 统计每个类型标签对应的电影量/条数/频数
    71.     data_lang_tj = pd.DataFrame(np.zeros([len(lang_list), 1]),
    72.                                 index=lang_list, columns=['tj'])  # 2列：标签，统计值tj
    73.     for i in data['language']:
    74.         for label in lang_list:
    75.             if str(i).__contains__(label):
    76.                 data_lang_tj.loc[label, 'tj'] += 1
    77.     # 将小类汇总为大类，并添加至统计df
    78.     chinese_fy = data_lang_tj.loc['湖南话', 'tj'] + data_lang_tj.loc['北京话', 'tj']
    79.     # print(chinese_fy)
    80.     data_lang_tj.loc['中国方言', 'tj'] = chinese_fy
    81.     return data_lang_tj
    84. def averge_votes():
    85.     return pd.read_csv(file, usecols=['average', 'votes', 'title'])
    88. def genre_tj():
    89.     # 读取数据
    90.     data = pd.read_csv(file)
    91.     # 获取所有类型：提取每一行的genre元素 -> 新的列表 genre_list -> 去重
    92.     data['genre'] = data['genre'].str.strip('[')
    93.     data['genre'] = data['genre'].str.strip(']')
    94.     data['genre'] = data['genre'].fillna(value='')
    95.     genre_list = []
    96.     for g in data['genre']:
    97.         g_list = g.split(', ')
    98.         for label in g_list:  # 123,2,3 -> 1,2,3
    99.             genre_list.append(label)
    100.     g_list = list(set(genre_list))
    101.     g_list.remove('')
    102.     # 统计每个类型标签对应的电影量/条数/频数
    103.     data_genre_tj = pd.DataFrame(np.zeros([len(g_list), 1]),
    104.                                  index=g_list, columns=['tj'])  # 2列：标签，统计值tj
    105.     for i in data['genre']:
    106.         for label in g_list:
    107.             if str(i).__contains__(label):
    108.                 data_genre_tj.loc[label, 'tj'] += 1
    109.     return data_genre_tj
    112. def genre_rates_tj(x):
    113.     '''
    114.     :param x: 前X为
    115.     :return: 排名前X位的电影类型标签，对应的评分均值数据
    116.     '''
    117.     # 电影类型(x)：6个标签，电影数量最多的
    118.     # 评分数据(y)：2，3，4，5，6，7，8，9，10
    119.     # 电影数量(值）
    120.     genre_list = genre_tj().sort_values('tj', ascending=False) \
    121.         .head(x).index.tolist()
    122.     rate_list = [2, 3, 4, 5, 6, 7, 8, 9, 10]
    123.     data = pd.read_csv(file, usecols=['genre', 'average'])
    124.     # 统计满足类型与评分区间的数据
    125.     data_genre_tj = pd.DataFrame(np.zeros([len(rate_list), len(genre_list)]),
    126.                                  index=rate_list, columns=genre_list)
    127.     for g in genre_list:  # 循环遍历6个变迁元素
    128.         for rate in rate_list:  # 循环遍历2-10评分数值
    129.             for i, r in zip(data['genre'], data['average']):
    130.                 if str(i).__contains__(g) and rate <= r < rate + 1:
    131.                     data_genre_tj.loc[rate, g] += 1
    132.     return data_genre_tj
    135. def rate_tj_by_year(year_list):
    136.     data = pd.read_csv(file,
    137.                        usecols=['title', 'average', 'release_date'])
    138.     data = data.set_index(pd.to_datetime(data['release_date']))
    139.     tj = []
    140.     for year in year_list:
    141.         tj.append(data[year]['average'].tolist())
    142.     return tj
    145. if __name__ == '__main__':
    146.     # print(movie_year_amount_tj())
    147.     # print(genre_rates_tj(6))
    148.     print(rate_tj_by_year(['2015', '2016']))