3. 在整个DataFrame上操作

  1.  In[18]: pd.options.display.max_rows = 8
  2.          movie = pd.read_csv('data/movie.csv')
  3.          #  打印行数和列数
  4.          movie.shape 
  5. Out[18]: (4916, 28)
  1. #  打印数据的个数
  2.  In[19]: movie.size
  3. Out[19]: 137648
  1. #  该数据集的维度
  2.  In[20]: movie.ndim
  3. Out[20]: 2
  1. #  该数据集的长度
  2.  In[21]: len(movie)
  3. Out[21]: 4916
  1. #  各个列的值的个数
  2.  In[22]: movie.count()
  3. Out[22]: color                     4897
  4.          director_name             4814
  5.          num_critic_for_reviews    4867
  6.          duration                  4901
  7.                                    ... 
  8.          actor_2_facebook_likes    4903
  9.          imdb_score                4916
  10.          aspect_ratio              4590
  11.          movie_facebook_likes      4916
  12.          Length: 28, dtype: int64
  1. #  各列的最小值
  2.  In[23]: movie.min()
  3. Out[23]: num_critic_for_reviews     1.00
  4.          duration                   7.00
  5.          director_facebook_likes    0.00
  6.          actor_3_facebook_likes     0.00
  7.                                     ... 
  8.          actor_2_facebook_likes     0.00
  9.          imdb_score                 1.60
  10.          aspect_ratio               1.18
  11.          movie_facebook_likes       0.00
  12.          Length: 16, dtype: float64
  1. #  打印描述信息
  2.  In[24]: movie.describe()
  3. Out[24]:

3. 在整个DataFrame上操作 - 图1

  1. #  使用percentiles参数指定分位数
  2.  In[25]: pd.options.display.max_rows = 10
  3.  In[26]: movie.describe(percentiles=[.01, .3, .99])
  4. Out[26]:

3. 在整个DataFrame上操作 - 图2

  1. #  打印各列空值的个数
  2.  In[27]: pd.options.display.max_rows = 8
  3.  In[28]: movie.isnull().sum()
  4. Out[28]: color                      19
  5.          director_name             102
  6.          num_critic_for_reviews     49
  7.          duration                   15
  8.                                    ... 
  9.          actor_2_facebook_likes     13
  10.          imdb_score                  0
  11.          aspect_ratio              326
  12.          movie_facebook_likes        0
  13.          Length: 28, dtype: int64

更多

  1. #  设定skipna=False，没有缺失值的数值列才会计算结果
  2.  In[29]: movie.min(skipna=False)
  3. Out[29]: num_critic_for_reviews     NaN
  4.          duration                   NaN
  5.          director_facebook_likes    NaN
  6.          actor_3_facebook_likes     NaN
  7.                                     ... 
  8.          actor_2_facebook_likes     NaN
  9.          imdb_score                 1.6
  10.          aspect_ratio               NaN
  11.          movie_facebook_likes       0.0
  12.          Length: 16, dtype: float64