CODEWITHSUNDEEP
pythonpython-3.xpandasdataframe
Sara Kerrigan
Sara Kerrigan

Reputation: 185

How to multiply certain values of a column by a constant?

I have a dataframe that looks like this:

     product     score
0    1179160  0.424654
1    1066490  0.424509
2    1148126  0.422207
3    1069104  0.420455
4    1069105  0.414603
..       ...       ...
491  1160330  0.168784
492  1069098  0.168749
493  1077784  0.168738
494  1193369  0.168703
495  1179741  0.168684

what I'm trying to achieve is to multiply certain score values corresponding to specific products by a constant. I have the products target of this multiplication in a list like this: [1069104, 1069105] (this is just a simplified example, in reality it would be more than two products) and my goal is to obtain this:

Multiply scores corresponding to products 1069104 and 1069105 by 10:

     product     score
0    1179160  0.424654
1    1066490  0.424509
2    1148126  0.422207
3    1069104  4.204550
4    1069105  4.146030
..       ...       ...
491  1160330  0.168784
492  1069098  0.168749
493  1077784  0.168738
494  1193369  0.168703
495  1179741  0.168684

I know that exists DataFrame.multiply but checking the examples it works for full columns, and I just one to change those specific values.

Upvotes: 6

Views: 2673

Answers (4)

Douw Marx
Douw Marx

Reputation: 31

import pandas as pd

# Make data
df = pd.DataFrame({'product': [2, 4, 8, 0],
                   'score': [0.2, 0.1, 0.2, 0.1]})

# List of products to change
product_list = [8,4]

for product_no in product_list:                    # Loop though product numbers
    product = df.loc[df['product'] == product_no]  # Find the product corresponding to given product number
    score = product["score"].values[0]             # Extract the score for the given product number

    print(score*10)                                # Modify score as you please

    df.at[product.index[0], "score"] = score*10    # If you want to modify the dataframe with the new value

print(df)

Upvotes: 0

Angad Gupta
Angad Gupta

Reputation: 101

We can use if condition with for loop to multiply the specific values

for i in range (len(df)):
    `if((df[colun]==1069104) or (df[colun]==1069105)):
          df[score]=df[score]*10

Upvotes: 0

Quang Hoang
Quang Hoang

Reputation: 150745

You can use boolean indexing and isin:

prod_list =  [1069104, 1069105]
df.loc[df['product'].isin(prod_list), 'score'] *= 10

Upvotes: 5

Tom Ron
Tom Ron

Reputation: 6181

You can use loc functionality (documentation)-

df = pd.DataFrame([[1, 'a'], [2, 'a'], [3, 'b']], columns=['num', 'c'])
df.loc[df.c == 'a', 'num'] =  df.num * 10

Upvotes: 1

Related Questions