熊猫 DataFrame 中如何将 Float 转换为 Datetime?
原文:https://www.geeksforgeeks.org/如何将浮点转换为熊猫中的日期时间-dataframe/
Pandas Dataframe 提供了更改列值数据类型的自由。我们可以将它们从整数更改为浮点类型,从整数更改为日期时间,从字符串更改为整数,从浮点更改为日期时间等。为了将浮点转换为日期时间,我们使用T1【pandas.to_DateTime()函数并遵循语法使用 :
语法:pandas.to_datetime (arg,errors='raise ',dayfirst=False,yearfirst=False,utc=None,box=True,format=None,exact=True,unit=None,infer_datetime_format = False,origin='unix ',cache=False)
示例 1: 使用 pandas.to_datetime() 将一列从浮点转换为“yyymmdd”格式
蟒蛇 3
# importing pandas library
import pandas as pd
# Initializing the nested list
# with Data set
player_list = [[20200112.0,'Mathematics'],
[20200114.0,'English'],
[20200116.0,'Physics'],
[20200119.0,'Chemistry'],
[20200121.0,'French'],
[20200124.0,'Biology'],
[20200129.0,'Sanskrit']]
# creating a pandas dataframe
df = pd.DataFrame(player_list,columns=['Dates','Test'])
# printing dataframe
print(df)
print()
# checking the type
print(df.dtypes)
输出:
更改数据类型后。
蟒蛇 3
# converting the float to datetime format
df['Dates'] = pd.to_datetime(df['Dates'], format='%Y%m%d')
# printing dataframe
print(df)
print()
print(df.dtypes)
输出:
在上例中,我们将列“date”的数据类型从“ float64 更改为“datetime 64【ns】”类型。
例 2: 如果数据帧列是 yymmdd 格式,我们必须将其转换为yymmdd格式
蟒蛇 3
# importing pandas library
import pandas as pd
# Initializing the nested list with
# Data set
player_list = [[180112.0,'Mathematics'],
[180114.0,'English'],
[180116.0,'Physics'],
[180119.0,'Chemistry'],
[180121.0,'French'],
[180124.0,'Biology'],
[180129.0,'Sanskrit']]
# creating a pandas dataframe
df = pd.DataFrame(player_list,columns=['Dates','Test'])
# printing dataframe
print(df)
print()
# checking the type
print(df.dtypes)
输出:
更改数据类型后。
蟒蛇 3
# converting the float to datetime format
df['Dates'] = pd.to_datetime(df['Dates'], format='%y%m%d')
# printing dataframe
print(df)
print()
print(df.dtypes)
输出:
在上例中,我们将列“date”的数据类型从“ float64 更改为“datetime 64【ns】,格式从“ yymmdd 更改为“yymmdd”。
示例 3: 当我们必须将浮动列转换为日期和时间格式时
蟒蛇 3
# importing pandas library
import pandas as pd
# Initializing the nested list with Data set
player_list = [[20200112082520.0,'Mathematics'],
[20200114085020.0,'English'],
[20200116093529.0,'Physics'],
[20200119101530.0,'Chemistry'],
[20200121104060.0,'French'],
[20200124113541.0,'Biology'],
[20200129125023.0,'Sanskrit']]
# creating a pandas dataframe
df = pd.DataFrame(player_list,columns=['Dates','Test'])
# printing dataframe
print(df)
print()
# checking the type
print(df.dtypes)
输出:
更改数据类型后。
蟒蛇 3
# converting the float to datetime format
df['Dates'] = pd.to_datetime(df['Dates'], format='%Y%m%d%H%M%S')
# printing dataframe
print(df)
print()
print(df.dtypes)
输出:
在上面的示例中,我们将列“Date”的数据类型从“ float64 更改为“datetime 64【ns】,并将格式更改为 Date 和 Time
示例 4: 使用 pandas.to_datetime() 将多个列从 float 转换为yyyymmdd格式
蟒蛇 3
# importing pandas library
import pandas as pd
# Initializing the nested list with Data set
player_list = [[20200112.0,'Mathematics',20200113.0],
[20200114.0,'English',20200115.0],
[20200116.0,'Physics',20200117.0],
[20200119.0,'Chemistry',20200120.0],
[20200121.0,'French',20200122.0],
[20200124.0,'Biology',20200125.0],
[20200129.0,'Sanskrit',20200130.0]]
# creating a pandas dataframe
df = pd.DataFrame(player_list,columns=['Starting_Date','Test','Ending_Date'])
# printing dataframe
print(df)
print()
# checking the type
print(df.dtypes)