7. 串联Series方法

  1. #  value_counts().head(3)，计数，查看前三
  2.  In[76]: movie = pd.read_csv('data/movie.csv')
  3.          actor_1_fb_likes = movie['actor_1_facebook_likes']
  4.          director = movie['director_name']
  5.  In[77]: director.value_counts().head(3)
  6. Out[77]: Steven Spielberg    26
  7.          Woody Allen         22
  8.          Clint Eastwood      20
  9.          Name: director_name, dtype: int64
  1. #  统计缺失值的数量
  2.  In[78]: actor_1_fb_likes.isnull().sum()
  3. Out[78]: 7
  1. #  actor_1_fb_likes的数据类型
  2.  In[79]: actor_1_fb_likes.dtype
  3. Out[79]: dtype('float64')
  1. #  缺失值填充为0、转换为整型、查看前五
  2.  In[80]: actor_1_fb_likes.fillna(0)\
  3.                 .astype(int)\
  4.                 .head()
  5. Out[80]: 0     1000
  6.          1    40000
  7.          2    11000
  8.          3    27000
  9.          4      131
  10.          Name: actor_1_facebook_likes, dtype: int64

更多

  1. #  缺失值的比例
  2.  In[81]: actor_1_fb_likes.isnull().mean()
  3. Out[81]: 0.0014239218877135883
  1. #  使用括号串联
  2.  In[82]: (actor_1_fb_likes.fillna(0)
  3.                  .astype(int)
  4.                  .head())
  5. Out[82]: 0     1000
  6.          1    40000
  7.          2    11000
  8.          3    27000
  9.          4      131
  10.          Name: actor_1_facebook_likes, dtype: int64