将元组列表列表转换为pandas datafram

Question

我正在尝试将元组列表转换为pandas数据帧，但不知道如何实现。我的地址结构如下：

addresses = [
 [('the vicars inn', 'house'), ('68', 'house_number'), ('church lane', 'road'), ('arlesey', 'city'), ('beds', 'house')],
 [('the old oak', 'house'), ('85', 'house_number'), ('church lane', 'road'), ('arlesey', 'city'), ('beds', 'house')],
 [('adj', 'road'), ('85', 'house_number'), ('high street', 'road'), ('arlesey', 'city'), ('beds', 'house')],
 [('arlesey community centre', 'house'), ('high street', 'road'), ('arlesey', 'city'), ('beds', 'house')],
 [('arlesey community centre', 'house'), ('high street', 'road'), ('arlesey', 'city'), ('beds', 'house')]
]

理想情况下，我需要一个数据帧：

^{pr2}$

到目前为止，我所做的只是改变一下局面，但并没有产生预期的结果：

pd.DataFrame.from_records(addresses[0]).pivot(columns=1, values=0)

有没有人对实现我理想数据帧的方法有任何指导？在

山姆

Answer 1

您可以将每条记录转换为字典，然后使用DataFrame.from_records：

pd.DataFrame.from_records([{k: v for v, k in row} for row in addresses])

#      city house   house_number    road
#0  arlesey beds              68    church lane
#1  arlesey beds              85    church lane
#2  arlesey beds              85    high street
#3  arlesey beds             NaN    high street
#4  arlesey beds             NaN    high street