从numpy loadtxt()获取日期列 [英] Get date column from numpy loadtxt()

问题描述

我有一个包含下表的文本文件.

I have a text file which contains following table.

Day  Month Year Avg Power
01     01  2000 30 
02     01  2000 41
04     01  2000 55
05     01  2000 78
06     01  2000 134
07     01  2000 42  

我想将日"，月"和年"列加载到单个datetime值中.为此，请按照以下步骤操作.但是代码无法按我期望的那样工作.

I want to load the Day,month and year columns into single datetime value. To do that followed following steps. But the code doesn't work what I expect.

from numpy import loadtxt
import datetime

def date_converter(x,y,z):
    date = "{},{},{}".format(x,y,z)
    return datetime.datetime.strptime(date,r"%d,%m,%Y")

data3 = loadtxt('complex_data_file.txt',dtype=int, usecols=(0,1,2,4),
                converters={(0,1,2):date_converter,3:int})

要达到我的要求我该怎么做?

What I have to do to achieve my requirement?

推荐答案

我将Pandas模块用于此任务:

I'd use Pandas module for this task:

In [228]: df = pd.read_csv(fn, usecols=[0,1,2,4], parse_dates={'Date':[2,1,0]})

In [229]: df
Out[229]:
        Date  Avg Power
0 2000-01-01         30
1 2000-01-02         41
2 2000-01-04         55
3 2000-01-05         78
4 2000-01-06        134
5 2000-01-07         42

In [230]: df.dtypes
Out[230]:
Date         datetime64[ns]
Avg Power             int64
dtype: object

将其转换为Numpy数组也很容易:

it's also very easy to convert it to a Numpy array:

In [231]: df.values
Out[231]:
array([[Timestamp('2000-01-01 00:00:00'), 30],
       [Timestamp('2000-01-02 00:00:00'), 41],
       [Timestamp('2000-01-04 00:00:00'), 55],
       [Timestamp('2000-01-05 00:00:00'), 78],
       [Timestamp('2000-01-06 00:00:00'), 134],
       [Timestamp('2000-01-07 00:00:00'), 42]], dtype=object)

这篇关于从numpy loadtxt()获取日期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！