CODEWITHSUNDEEP
pythonpandas
ajsp
ajsp

Reputation: 2670

Find consecutively larger Values in column

Say i have a data frame:

df = pd.DataFrame({'A':[5,4,7,8,1,2,3,4,5,7,8,9],'B':[1,2,2,2,2,5,9,8,8,10,11,10]})
print df

    A   B
0   5   1
1   4   2
2   7   2
3   8   2
4   1   2
5   2   5
6   3   9
7   4   8
8   5   8
9   7  10
10  8  11
11  9  10

And I want to find only the consecutively higher values in df.A, therefore df would become: 

    A   B
0   5   1
2   7   2
3   8   2
11  9  10

What is the best(read fastest) way forward here; I have tried something quite complicated but alas, it's actually slower than looping thorough the whole frame.

Thanks.

Upvotes: 0

Views: 52

Answers (1)

DYZ
DYZ

Reputation: 57033

This solution works only if B grows monotonously:

df.cummax().drop_duplicates('A')
#    A   B 
#0   5   1 
#2   7   2 
#3   8   2 
#11  9  10

This one is more general:

df['C'] = df['A'].cummax()
df.drop_duplicates('C')[['A','B']]

Upvotes: 1

Related Questions