3. 用布尔索引过滤

  1. #  读取movie数据集，创建布尔条件
  2.  In[15]: movie = pd.read_csv('data/movie.csv', index_col='movie_title')
  4.          crit_a1 = movie.imdb_score > 8
  5.          crit_a2 = movie.content_rating == 'PG-13'
  6.          crit_a3 = (movie.title_year < 2000) | (movie.title_year > 2009)
  7.          final_crit_a = crit_a1 & crit_a2 & crit_a3
  8. #  创建第二个布尔条件
  9.  In[16]: crit_b1 = movie.imdb_score < 5
  10.          crit_b2 = movie.content_rating == 'R'
  11.          crit_b3 = (movie.title_year >= 2000) & (movie.title_year <= 2010)
  12.          final_crit_b = crit_b1 & crit_b2 & crit_b3
  13. #  将这两个条件用或运算合并起来
  14.  In[17]: final_crit_all = final_crit_a | final_crit_b
  15.          final_crit_all.head()
  16. Out[17]: movie_title
  17.          Avatar                                        False
  18.          Pirates of the Caribbean: At World's End      False
  19.          Spectre                                       False
  20.          The Dark Knight Rises                          True
  21.          Star Wars: Episode VII - The Force Awakens    False
  22.          dtype: bool
  1. #  用最终的布尔条件过滤数据
  2.  In[18]: movie[final_crit_all].head()
  3. Out[18]:

3. 用布尔索引过滤 - 图1

  1. #  使用loc，对指定的列做过滤操作，可以清楚地看到过滤是否起作用
  2.  In[19]: cols = ['imdb_score', 'content_rating', 'title_year']
  3.          movie_filtered = movie.loc[final_crit_all, cols]
  4.          movie_filtered.head(10)
  5. Out[19]:

3. 用布尔索引过滤 - 图2

更多

  1. #  用一个长布尔表达式代替前面由短表达式生成的布尔条件
  2.  In[21]: final_crit_a2 = (movie.imdb_score > 8) & \
  3.                          (movie.content_rating == 'PG-13') & \
  4.                          ((movie.title_year < 2000) | (movie.title_year > 2009))
  5.          final_crit_a2.equals(final_crit_a)
  6. Out[21]:

3. 用布尔索引过滤 - 图3