所以我有一个数据帧,我正在分组,然后应用一个函数。现在我想检查帧中的每一行检查数据帧中的剩余行,如果它匹配某些条件,我想将它们添加到具有某种标记的不同数据帧并从原始中删除它们。如果它没有通过条件我将行保留在那里并继续前进到下一行。
e.g
time status number action fname lname
0 10.30 Active 2 0 Adrian Peter
1 11.01 Active 3 2 Peter Thomas
2 11.05 Passive 2 0 Thomas Adrian
3 11.07 Passive 2 1 Jen Anniston
所以我做了像
这样的事情 df.groupby(status).apply(f)
def f(x):
I want to perform some tasks here and with the remaining dataframe
i want to see if index 0 has similar number and action in the
remaining data frame. If true i want to put this in a different dataframe and tag it and remove the pair from the origial df.
I want to then move on to the next index and do the same. If false after looking at all the data in the frame i want to delete this from the original df too
答案 0 :(得分:1)
如果你想要的函数(f)有副作用,我会使用df.iterrows()并在python中编写函数。
for index, row in df.iterrows():
# Do stuff
您还可以使用布尔值创建一个标记列来评估您的条件,然后弹出将该值设置为true的所有行:
df['tagged'] = df.apply(lambda row: <<condition goes here>>, axis=1)
tagged_rows = df[df['tagged'] == True]
df = df[df['tagged'] != True]
(不是100%确定语法,手头没有解释器)