Question

我有多个标头的csv文件：

Year', 'Month', 'Day', 'Hour', 'Minute' --other headers--

我想用datetime缩小这些尺寸。现在我正在做：

date = datetime(year=int(d[0]), month=int(d[1]), day=int(d[2]]), hour=int(d[3]]), minute=int(d[4]]))

但是，如何实际删除这些列并将日期列放在所有其他列之前？我最终希望“日期”成为熊猫数据框中的索引。

更新：我的一个CSV的前五行是：

"Year","Month","Day","Hour","Minute","Direct","Diffuse","D_Global","D_IR","U_Global","U_IR","Zenith"
2015,4,1,0,1,17.59,13.27,-999.00,158.94,10.75,178.97,88.700
2015,4,1,0,2,20.23,12.67,-999.00,159.45,10.68,178.99,88.726
2015,4,1,0,3,20.90,12.42,-999.00,159.77,10.55,179.03,88.752
2015,4,1,0,4,16.38,11.93,-999.00,160.68,10.47,179.45,88.777

我想

"Date","Direct","Diffuse","D_Global","D_IR","U_Global","U_IR","Zenith"
DATETIMEOBJECT,17.59,13.27,-999.00,158.94,10.75,178.97,88.700
DATETIMEOBJECT,20.23,12.67,-999.00,159.45,10.68,178.99,88.726
DATETIMEOBJECT,20.90,12.42,-999.00,159.77,10.55,179.03,88.752
DATETIMEOBJECT,16.38,11.93,-999.00,160.68,10.47,179.45,88.777

Answer 1

您可以将parse_dates参数指定为与日期时间相关的列的索引，并编写自定义的date_parser函数以将多列转换为日期时间列。如果您希望它成为索引，请指定index_col：

import pandas as pd   
from io import StringIO   
pd.read_csv(StringIO("""
"Year","Month","Day","Hour","Minute","Direct","Diffuse","D_Global","D_IR","U_Global","U_IR","Zenith"
2015,4,1,0,1,17.59,13.27,-999.00,158.94,10.75,178.97,88.700
2015,4,1,0,2,20.23,12.67,-999.00,159.45,10.68,178.99,88.726
2015,4,1,0,3,20.90,12.42,-999.00,159.77,10.55,179.03,88.752
2015,4,1,0,4,16.38,11.93,-999.00,160.68,10.47,179.45,88.777"""), 
            sep = ",", 
            parse_dates={'Date': [0,1,2,3,4]}, 
            date_parser = lambda x: pd.to_datetime(x, format="%Y %m %d %H %M"), 
            index_col = ['Date'])

更新： date_parser函数的工作方式符合pd.read_csv() docs：

1）传递一个或多个阵列（由parse_dates定义）作为参数; 2）连接（行方式）从parse_dates定义的列到单个数组的字符串值通过那个; 3）使用一个或多个为每行调用date_parser一次字符串（对应于parse_dates定义的列）作为参数。

根据自定义日期解析器函数的编写方式，有多种方法可以使用此参数，在上面给出的情况下，例如，使用第二个选项，对于每一行，列0-4连接为字符串由空格分隔并传递给pd.to_datetime()函数，因此格式为%Y %m %d %H %M。

Python用datetime替换多个日期列

1 个答案: