CODEWITHSUNDEEP
pandas
Sht
Sht

Reputation: 189

How to split pandas columns according to some rule, and sum up the values

I have columns that are values 2/300/3 like this, but I want some of this value, in this case, 2+300+3=305. In order to achieve this goal I am using the command

for i in range(len(df.col)):
    df.col.iloc[i]=sum([float(j) for j in df.col.iloc[i].split("|")])

The length of this data frame is big, that why the above command is very time-consuming. Does there exist another way to do the same operation without loop? NOTE: the number of "/" is not fixed and in some cases it does not exist

Upvotes: 0

Views: 129

Answers (2)

Yefet
Yefet

Reputation: 2086

replace and eval can do the job

import pandas as pd

data = ["2/300/3" , "5/300/3" , "3 50 3"]

df =pd.DataFrame(data, columns=["data"])

df["total_sum"] = df["data"].str.replace('[/ |" "]','+').apply(pd.eval)

Output:

    data    total_sum
0   2/300/3 305
1   5/300/3 308
2   3 50 3  56

Upvotes: 1

Nk03
Nk03

Reputation: 14949

you can also use str.split and apply

import pandas as pd
data = ["2/300/3" , "5/300/3" , "3/50/3"]
df =pd.DataFrame(data, columns=["data"])
df.data = df.data.str.split('/').apply(lambda x: sum(map(int,x)))

Upvotes: 0

Related Questions