访问没有名称的列并从dataframe中删除某些数据

Question

我有一个dataframe，它有14列，其中没有任何标题。第一列是date，我需要删除从现在起超过25个月且不包含任何日期的所有行

                        44.93442    -79.37061   Tow 36  45.06541    -79.43384   R103    2053    WL7 CCG     
                        44.96092    -78.39428   Flatbed Tow 56  0   0   P201    3040    SHOP    CCG     
                        45.05056    -78.50632   Jump Start/Battery Test 30  45.12152    -78.61311   P201    1426    FB1 CCG     
                        45.15709    -77.91703   Winch/Extrication   20  0   0   K143    957 SHOP    CCG     
                        45.24038    -81.64227   Tow 36  45.25674    -81.65996   W444    1597    WL1 CCG     
                        45.32589    -78.98001   Winch/Extrication   79  0   0   R105    43  SHOP    CCG     
                        45.33402    -79.20586   Tow 38  45.32871    -79.21218   R103    226 WL10    CCG     
                        46.2062     -82.47623   Tow 68  46.18153    -82.96333   R812    588 FM1 CCG     
                        46.50164    -84.28004   Winch/Extrication   36  46.53398    -84.35199   R829    10  WL1 CCG     
                        46.50694    -84.32854   No service  27  46.53578    -84.38477   R829    124 WL1 CCG     
                        46.51246    -84.33669   No service  728 46.50897    -84.3295    R829    123 FB1 CCG     
                        46.52354    -84.38171   Flatbed Tow 66  46.51035    -84.25437   R829    49  FB2 CCG     
2017-11-11  00:03:10    43.68109    -79.53289   Tow 33  43.54982    -79.71832   C1  2   WL2 CCG     00:05:50
2017-11-11  00:04:09    43.91352    -78.9673    Tow 24  43.93242    -78.86989   C207    4   M255    CCG     00:04:11
2017-11-11  00:05:09    42.93152    -81.19933   Tow 161 42.91458    -81.2216    G72 5   FB6 CCG P5  00:16:21
2017-11-11  00:05:14    44.26861    -76.57645   Jump Start/Battery Test 21  44.26594    -76.49862   K112    6   LS1 CCG     00:12:23
2017-11-11  00:07:09    43.30024    -79.90106   Flatbed Tow 53  43.30506    -79.90795   H935    8   350 CCG P5  00:14:35
2017-11-11  00:07:14    43.71246    -79.55377   Tow 40  43.7081 -79.5536    C100    9   WL25    CCG P2  00:07:22

到目前为止，我已经试过了，但还没有成功

import pandas as pd
from datetime import datetime,timedelta

#find the 25 month prior month from now 

N = 761
today = datetime.now()
lastMonth = today-timedelta(days=N)
month24= lastMonth.strftime("%Y-%m")
print (month24)
data = pd.read_csv("E:/ERS_DATA_TEMP.txt",delimiter='\t', dtype=str)

#access the dataframe and column

x=data.iloc[:,0]
print (x)
#remove data 
data.drop(data[data.iloc[:,0]],[np.NaN]>=month24)
y=len(data.index)

Answer 1

读取不带标题的数据

df = pd.read_csv("E:/ERS_DATA_TEMP.txt",delimiter='\t', dtype=str, header=None)

没有时间就删除所有行

df = df[df[0].apply(pd.to_datetime, errors='coerce').notnull()]

最后切上日期

df = df[df[0] > month24]

Answer 2

您可以使用iloc访问器对列进行操作：

# convert month to days
n_days = 25 * 30

# both will return a boolean series
t1 = df.iloc[:,0].apply(lambda x: (x - pd.to_datetime('today')).days).gt(n_days)
t2 = df.iloc[0].isna()

# remove unwanted dates
df1 = df.loc[t1 & t2]

Sample Data

period = pd.date_range(start='20130101', freq='M', periods=1000)
df = pd.DataFrame({'period': period})