如何将具有相似名称的列的 pandas dataframe 转换为行？

Question

例如，我有下面的 dataframe

我想变成这样的东西：

Timetsamp     | invoice_number | invoice_amount

10/14/2019    |   20565116     |  101.09

10/14/2019    |   20617023     |  505.57

10/15/2019    |   20600496     |  503.2

10/15/2019    |   20582306     |  501.11

Answer 1

# timestamp dataframe
timestamp = df[['Timestamp']]

# stores the new dataframe with timestamp and invoices
new_df = pd.DataFrame():

# iterate through the cols in original dataframe in steps of 2
for c in range(1,len(df.columns),2):

    # get the column invoice number and amount
    temp = df[[df.columns[c],df.columns[c+1]]
    # concat with the timestamp
    temp = pd.concat([timestamp, temp], axis = 1)
    # make a bigger dataframe
    new_df = pd.concat([new_df, temp], axis = 0, ignore_index = True)

Answer 2

像这样的东西会起作用：

group = df.set_index('timestamp').T.groupby(df.columns[1:])
df = pd.concat([grp.T.stack().droplevel(1).rename(idx).to_frame() for idx, grp in group], 1)