如何根据日期创建单独的跟踪列? [英] How to create a separate tracking column based on dates?

问题描述

我有一个数据框，df:

I have a dataframe, df:

df:
            val
date
2012-01-01  4.2      
2012-01-02  3.7
2012-01-03  6.2
2012-01-04  1.2
2012-01-05  2.4
2012-01-06  2.3

我要创建的列是从 0 开始指定日期并相应地填充该列(假设本例中的日期为 2012-01-04):

What I want to create is a column that starts at 0 for a specified date and fills in the column accordingly (assume the date in this case is 2012-01-04):

df2:
            val  tracking
date
2012-01-01  4.2  -3
2012-01-02  3.7  -2
2012-01-03  6.2  -1
2012-01-04  1.2  0
2012-01-05  2.4  1
2012-01-06  2.3  2

我尝试使用 np.arange()，但无法以我需要的行为中心.日期列设置为索引(pandas df).

I tried using np.arange() but was having trouble centering on the row I needed. The date column is set up as an index (pandas df).

谢谢.

推荐答案

我认为最简单的方法是分两部分进行:

I think the easiest way to do this is to do it in two parts:

df['tracking'] = pd.np.arange(len(df))

In [12]: df
Out[12]: 
            val  tracking
date                     
2012-01-01  4.2         0
2012-01-02  3.7         1
2012-01-03  6.2         2
2012-01-04  1.2         3
2012-01-05  2.4         4
2012-01-06  2.3         5

df['tracking'] -= df.ix['2012-01-04']['tracking']

In [14]: df
Out[14]: 
            val  tracking
date                     
2012-01-01  4.2        -3
2012-01-02  3.7        -2
2012-01-03  6.2        -1
2012-01-04  1.2         0
2012-01-05  2.4         1
2012-01-06  2.3         2

这篇关于如何根据日期创建单独的跟踪列?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！