我想在我的數據幀中將1-May-19和5/1/2019轉換為1/5/2019

Question

我有2列日期格式：5/1/2019和1-May-19。 我希望他們在1/5/2019格式。

這里提到的代碼似乎沒有改變1列的格式，我需要將這兩列轉換為格式1/5/2019

df['billing_start_date'] = (pd.to_datetime(df['billing_start_date'], format='%Y/%m/%d').dt.strftime('%m/%d/%Y'))

Answer 1

使用帶有不同format的to_datetime的Series.dt.strftime ：

df = pd.DataFrame({"a":["5/1/2019"], "b":["1-May-19"]})

df['a'] = pd.to_datetime(df['a'], format='%m/%d/%Y').dt.strftime('%m/%d/%Y')
df['b'] = pd.to_datetime(df['b'], format='%d-%b-%y').dt.strftime('%m/%d/%Y')

print (df)
            a           b
0  05/01/2019  05/01/2019

Answer 2

# I guess you have two different columns with different dateformats as per the #  question , please check the below answer.

from datetime import datetime
df=pd.DataFrame({'start_date':['5/1/2019','7/11/2019','3/10/2019'],
                 'end_date':['1-May-19','2-Mar-19','11-May-19']})
df

    start_date  end_date
0   5/1/2019    1-May-19
1   7/11/2019   2-Mar-19
2   3/10/2019   11-May-19

def try_parsing_date(text):
    for fmt in ("%m/%d/%Y",'%d-%b-%y'):
        try:
            return datetime.strptime(text, fmt).strftime('%d/%m/%Y')
        except ValueError:
            pass
    raise ValueError('no valid date format found')

df['new_start_date']=df['start_date'].apply(try_parsing_date)
df['new_end_date']=df['end_date'].apply(try_parsing_date)
df

   start_date   end_date    new_start_date  new_end_date
0   5/1/2019    1-May-19    01/05/2019  01/05/2019
1   7/11/2019   2-Mar-19    11/07/2019  02/03/2019
2   3/10/2019   11-May-19   10/03/2019  11/05/2019

Answer 3

你有正確的方法，但執行錯誤

我們的開始df：

df = pd.DataFrame({"col_1":["5/1/2019"], "col_2":["1-May-19"]})

col_1       col_2
5/1/2019    1-May-19

以下是如何通過簡單的解決方案實現此目的：

df["col_1"] = pd.to_datetime(df["col_1"])
df["col_2"] = pd.to_datetime(df["col_1"])
df["col_1"] = df["col_1"].dt.strftime('%d/%m/%Y')
df["col_2"] = df["col_2"].dt.strftime('%d/%m/%Y')

這使 ：

col_1   col_2
01/05/2019  01/05/2019

或者用方法鏈接兩行

df["col_1"] = pd.to_datetime(df["col_1"]).dt.strftime('%d/%m/%Y')
df["col_2"] = pd.to_datetime(df["col_1"]).dt.strftime('%d/%m/%Y')