10. 用where方法保留Series

#  读取movie数据集，movie_title作为行索引，actor_1_facebook_likes列删除缺失值
 In[68]: movie = pd.read_csv('data/movie.csv', index_col='movie_title')
         fb_likes = movie['actor_1_facebook_likes'].dropna()
         fb_likes.head()
Out[68]: movie_title
         Avatar                                         1000.0
         Pirates of the Caribbean: At World's End      40000.0
         Spectre                                       11000.0
         The Dark Knight Rises                         27000.0
         Star Wars: Episode VII - The Force Awakens      131.0
         Name: actor_1_facebook_likes, dtype: float64

#  使用describe获得对数据的认知
 In[69]: fb_likes.describe(percentiles=[.1, .25, .5, .75, .9]).astype(int)
Out[69]: count      4909
         mean       6494
         std       15106
         min           0
         10%         240
         25%         607
         50%         982
         75%       11000
         90%       18000
         max      640000
         Name: actor_1_facebook_likes, dtype: int64

#  作用和前面相同（这里是作者代码弄乱了）
 In[70]: fb_likes.describe(percentiles=[.1,.25,.5,.75,.9])
Out[70]: count      4909.000000
         mean       6494.488491
         std       15106.986884
         min           0.000000
         10%         240.000000
         25%         607.000000
         50%         982.000000
         75%       11000.000000
         90%       18000.000000
         max      640000.000000
         Name: actor_1_facebook_likes, dtype: float64

#  画一张柱状图
 In[71]: fb_likes.hist()
Out[71]: <matplotlib.axes._subplots.AxesSubplot at 0x10f9fbe80>

#  检测小于20000个喜欢的的比例
 In[72]: criteria_high = fb_likes < 20000
         criteria_high.mean().round(2)
Out[71]: 0.91000000000000003

#  where条件可以返回一个同样大小的Series，但是所有False会被替换成缺失值
 In[73]: fb_likes.where(criteria_high).head()
Out[73]: movie_title
         Avatar                                         1000.0
         Pirates of the Caribbean: At World's End          NaN
         Spectre                                       11000.0
         The Dark Knight Rises                             NaN
         Star Wars: Episode VII - The Force Awakens      131.0
         Name: actor_1_facebook_likes, dtype: float64

#  第二个参数other，可以让你控制替换值
 In[74]: fb_likes.where(criteria_high, other=20000).head()
Out[74]: movie_title
         Avatar                                         1000.0
         Pirates of the Caribbean: At World's End      20000.0
         Spectre                                       11000.0
         The Dark Knight Rises                         20000.0
         Star Wars: Episode VII - The Force Awakens      131.0
         Name: actor_1_facebook_likes, dtype: float64

#  通过where条件，设定上下限的值
 In[75]: criteria_low = fb_likes > 300
         fb_likes_cap = fb_likes.where(criteria_high, other=20000)\
                                .where(criteria_low, 300)
         fb_likes_cap.head()
Out[75]: movie_title
         Avatar                                         1000.0
         Pirates of the Caribbean: At World's End      20000.0
         Spectre                                       11000.0
         The Dark Knight Rises                         20000.0
         Star Wars: Episode VII - The Force Awakens      300.0
         Name: actor_1_facebook_likes, dtype: float64

#  原始Series和修改过的Series的长度是一样的
 In[76]: len(fb_likes), len(fb_likes_cap)
Out[76]: (4909, 4909)

#  再做一张柱状图，效果好多了
 In[77]: fb_likes_cap.hist()
Out[77]: <matplotlib.axes._subplots.AxesSubplot at 0x10eeea8d0>

 In[78]: fb_likes_cap2 = fb_likes.clip(lower=300, upper=20000)
         fb_likes_cap2.equals(fb_likes_cap)
Out[78]: True