CODEWITHSUNDEEP
pythonpandasdataframe
Lucio Stortoni Ruiz
Lucio Stortoni Ruiz

Reputation: 33

Python Pandas change values from a condition

how are you? I'm pretty new at code, and I have this question:

I want to iterate through a column, and I want to change this column values based on a condition, in this case, I want to change very value from column 'a1': if the value contains the word 'Juancito' I want to change it to just 'Juancito'. The for loop works OK, but the value doesn't change in the end.

What I'm doing wrong?

import pandas as pd
inp = [{'a1':'Juancito 1'}, {'a1':'Juancito 2'}, {'a1':'Juancito 3'}]
df = pd.DataFrame(inp)

for i in df['a1']:
    if 'Juancito' in i:
        i = 'Juancito'
    else:
        pass
df.head()

Upvotes: 3

Views: 226

Answers (4)

Mayank Porwal
Mayank Porwal

Reputation: 34056

You don't need a for loop.

Just use numpy.where with Series.str.contains:

In [83]: import numpy as np

In [84]: df['a1'] = np.where(df['a1'].str.contains('Juancito'), 'Juancito', df['a1'])

In [85]: df
Out[85]: 
         a1
0  Juancito
1  Juancito
2  Juancito

Upvotes: 3

Epsi95
Epsi95

Reputation: 9047

As mentioned in the comment, incorporated the changes

import pandas as pd
inp = [{'a1':'Juancito 1'}, {'a1':'Juancito 2'}, {'a1':'Juancito 3'}]
df = pd.DataFrame(inp)

for index, i in enumerate(df['a1'])::
    if 'Juancito' == i:
        df['a1'].loc[index] = 'Juancito'
    else:
        pass
df.head()

Upvotes: 2

ahmadjanan
ahmadjanan

Reputation: 445

You are not setting the value of the dataframe element in your code. You are just assigning it temporarily to i. One approach would be to use numeric indexing where row number will change with i and column number would be the one you want to process.

Example:

import pandas as pd
inp = [{'a1':'Juancito 1'}, {'a1':'Juancito 2'}, {'a1':'Juancito 3'}]
df = pd.DataFrame(inp)

for i in range(len(df)):
    if 'Juancito' in df.iloc[i][0]:
        df.iloc[i][0] = 'Juancito'
    else:
        pass
df.head()

Upvotes: 2

dimay
dimay

Reputation: 2804

Try this:

import pandas as pd
inp = [{'a1':'Juancito 1'}, {'a1':'Juancito 2'}, {'a1':'Juancito 3'}]
df = pd.DataFrame(inp)

for i in range(df["a1"].shape[0]):
    print(i)
    if 'Juancito' in df["a1"][i]:
        df["a1"][i] = 'Juancito'
    else:
        pass
df.head()

Upvotes: 2

Related Questions