vault backup: 2024-03-17 18:58:39

Extras/Omnivore/数据处理过程.md

@@ -82,8 +82,6 @@ plt.show()
 
 ## 平均客单价
 
-为避免极端值影响，先按月份将所有数据分组，剔除前 1%和后 1%的订单后再计算平均客单价
-
 ```python
 # Attempting the analysis again with additional checks  
 import pandas as pd  
@@ -337,8 +335,6 @@ plt.show()
 
 ![image.png|600](https://image.kfdr.top/i/2024/03/16/65f5625b9cac2.png)
 
-考虑到表格中的预约类型分为马上出发和预约派车两种，这意味着实际业务的发生时间往往与系统记录的订单时间不匹配，因此将预约派车类型的订单全部剔除，只研究马上出发订单的时间分布
-
 ```python
 # Filter out booked departures to focus on immediate departures only  
 immediate_departures = data[data['预约类型'] == '马上出发']  
@@ -370,8 +366,6 @@ plt.show()
 
 ![image.png|600](https://image.kfdr.top/i/2024/03/16/65f5a29dc8dd5.png)
 
-更进一步，剔除掉疫情期间的所有业务，能够较为客观地反映现在的情况
-
 ```python
 # Filter for immediate departures after December 2022  
 immediate_departures_after_dec2022 = immediate_departures[immediate_departures['YearMonth'] > '2022-12']