run a loop inside another loop in pandas

Question

I have a sequential grouped table in Pandas.

I am trying to create a running sum within groups, conditional upon running sum can not be negative using the loop below I am able to create the same As you can see for the user_id 77558 the subtotal continues from user_id 223 how do I Fix this

user_id = [4705,4705,4705,4705,4705,223,223,223,223,223,223,223,77558,77558,77558,77558,77558,77558,77558,77558,77558,77558]
transaction_c= [1,2,3,4,5,1,2,3,4,5,6,7,1,2,3,4,5,6,7,8,9,10]
Credit_Debit = [75,-125,47,75,-122,50,50,100,-200,35,50,-15,100,27,27,-54,1000,-220,-220,-220,-220,1000,]

df = pd.DataFrame(list(zip(user_id,transaction_c,Credit_Debit)), columns =['user_id','transaction', 'Credit_Debit'])

#summary function 

lastvalue = 0
newtotal = []
for row in df['Credit_Debit']:
    thisvalue =  row + lastvalue
    if thisvalue < 0:
        thisvalue = 0
    newtotal.append( thisvalue )
    lastvalue = thisvalue
    
df['Balance']=pd.Series(newtotal, index=df.index)

output


+---------+-------------+--------------+----------+------------------+
| user_id | transaction | Credit_Debit | Balance  | Desired Balance  |
+---------+-------------+--------------+----------+------------------+
|    4705 |           1 |           75 |       75 |               75 |
|    4705 |           2 |         -125 |        0 |                0 |
|    4705 |           3 |           47 |       47 |               47 |
|    4705 |           4 |           75 |      122 |              122 |
|    4705 |           5 |         -122 |        0 |                0 |
|     223 |           1 |           50 |       50 |               50 |
|     223 |           2 |           50 |      100 |              100 |
|     223 |           3 |          100 |      200 |              200 |
|     223 |           4 |         -200 |        0 |                0 |
|     223 |           5 |           35 |       35 |               35 |
|     223 |           6 |           50 |       85 |               85 |
|     223 |           7 |          -15 |       70 |               70 |
|   77558 |           1 |          100 |      170 |              100 |
|   77558 |           2 |           27 |      197 |              127 |
|   77558 |           3 |           27 |      224 |              154 |
|   77558 |           4 |          -54 |      170 |              100 |
|   77558 |           5 |         1000 |     1170 |             1100 |
|   77558 |           6 |         -220 |      950 |              880 |
|   77558 |           7 |         -220 |      730 |              660 |
|   77558 |           8 |         -220 |      510 |              440 |
|   77558 |           9 |         -220 |      290 |              220 |
|   77558 |          10 |         1000 |     1290 |             1220 |
+---------+-------------+--------------+----------+------------------+

Appreciate all the help in doing resolving this . thanks :)

Rishabh Kumar · Accepted Answer · 2021-03-13 10:26:06Z

You can first group your dataframes using groupby and then apply your function (I named it "change"), on those individual groups.

import pandas as pd

user_id = [4705,4705,4705,4705,4705,223,223,223,223,223,223,223,77558,77558,77558,77558,77558,77558,77558,77558,77558,77558]
transaction_c= [1,2,3,4,5,1,2,3,4,5,6,7,1,2,3,4,5,6,7,8,9,10]
Credit_Debit = [75,-125,47,75,-122,50,50,100,-200,35,50,-15,100,27,27,-54,1000,-220,-220,-220,-220,1000,]

df = pd.DataFrame(list(zip(user_id,transaction_c,Credit_Debit)), columns =['user_id','transaction', 'Credit_Debit'])

#summary function 

def change(df):
    lastvalue = 0
    newtotal = []
    for row in df['Credit_Debit']:
        thisvalue =  row + lastvalue
        if thisvalue < 0:
            thisvalue = 0
        newtotal.append( thisvalue )
        lastvalue = thisvalue
    return pd.Series(newtotal)
    
df['Balance']= df.groupby('user_id',sort=False).apply(change).reset_index(drop=True)

print(df)

Output:

    user_id  transaction  Credit_Debit  Balance
0      4705            1            75       75
1      4705            2          -125        0
2      4705            3            47       47
3      4705            4            75      122
4      4705            5          -122        0
5       223            1            50       50
6       223            2            50      100
7       223            3           100      200
8       223            4          -200        0
9       223            5            35       35
10      223            6            50       85
11      223            7           -15       70
12    77558            1           100      100
13    77558            2            27      127
14    77558            3            27      154
15    77558            4           -54      100
16    77558            5          1000     1100
17    77558            6          -220      880
18    77558            7          -220      660
19    77558            8          -220      440
20    77558            9          -220      220
21    77558           10          1000     1220

Angry Coder · Accepted Answer · 2021-03-13 10:24:37Z

1

What's happening in your code is, that it's adding the last Transaction. Change Credit_Debit[11] = -85, then you will get your desired output. You have to reset your transaction for every unique ID.

LastTrans = 0
NewTotal = []
for i, col in zip(df['transaction'], df['Credit_Debit']):
    if i == 1:
        ThisValue = col
    else:
        ThisValue = col + LastTrans
        if ThisValue < 0:
            ThisValue = 0
    NewTotal.append(ThisValue)
    LastTrans = ThisValue

I hope this helps...

edited Mar 13, 2021 at 10:24

answered Mar 13, 2021 at 10:07

Angry Coder

822 silver badges8 bronze badges

1 Comment

Angry Coder Over a year ago

@AniruddhaDas I am trying to figure out. I'll post if I get the answer

Collectives™ on Stack Overflow

run a loop inside another loop in pandas

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related