DataFrame column values in specific format

Question

Here is my script ....

import pandas as pd

data = {'Col1': [132, 148, 149], 'Col2': [232, 248, 249], 'Col3': [312, 308, 309], 'Col4': [1500, 1550, 1600], 'Col5': [530, 590, 568]}
df = pd.DataFrame(data)
print(df)

def lrgst(df, cols, n):
    for key, col in df.set_index('Col4')[cols].items():
        x = col.nlargest(n)
        J=', '.join(f'{a}[{b}]' for a,b in zip(x.index, x))
        return J
print(f"{lrgst(df, ['Col1'], 2)}") 

With output as...

1600[149], 1550[148]

Now I want to create another function to get value of some special column (let's say Col3) in the same row in which col1 values are found in my function and want to keep them inside bracket of my above output. My output should be like this...

1600[149,309], 1550[148,308]

I want my new function something like this...

def lrgst(df, cols, sp_col, n):
    for key, col in df.set_index('Col4')[cols].items():
        x = col.nlargest(n)
        J=', '.join(f'{a}[{b},{c}]' for a,b,c in zip(x.index, x, idx.sp_col))
        return J
print(f"{lrgst(df, ['Col1'], ['Col3'], 2)}")

Plz may I have any help here ???

Answer 1

Please add:

sp_col_values = df.set_index('Col4')[sp_cols].loc[x.index]
J = ', '.join(f'{a}[{b},{c}]' for a, b, c in zip(x.index, x, sp_col_values.iloc[:, 0]))

All Function:

def lrgst(df, cols, sp_cols, n):
    results = []
    for key, col in df.set_index('Col4')[cols].items():
        x = col.nlargest(n)
        sp_col_values = df.set_index('Col4')[sp_cols].loc[x.index]
        J = ', '.join(f'{a}[{b},{c}]' for a, b, c in zip(x.index, x, sp_col_values.iloc[:, 0]))
        results.append(J)
    return results

print(', '.join(lrgst(df, ['Col1'], ['Col3'], 2)))

Output:

1600[149,309], 1550[148,308]

EDIT:

Calling function in the same way that u mentioned on the post: print(f"{lrgst(df, ['Col1'], ['Col3'], 2)}")

New function:

def lrgst(df, cols, sp_cols, n):
    result = []
    for key, col in df.set_index('Col4')[cols].items():
        x = col.nlargest(n)
        sp_col_values = df.set_index('Col4')[sp_cols].loc[x.index]
        J = ', '.join(f'{a}[{b},{c}]' for a, b, c in zip(x.index, x, sp_col_values.iloc[:, 0]))
        result.append(J)
    return ', '.join(result)

print(f"{lrgst(df, ['Col1'], ['Col3'], 2)}")

Another way:

def lrgst(df, cols, sp_cols, n):
    return ', '.join(
        f'{a}[{b},{c}]'
        for key, col in df.set_index('Col4')[cols].items()
        for a, b, c in zip(col.nlargest(n).index, col.nlargest(n).values, df.set_index('Col4')[sp_cols].loc[col.nlargest(n).index][sp_cols[0]])
    )

print(f"{lrgst(df, ['Col1'], ['Col3'], 2)}")

Answer 2

Here's an updated version of your function lrgst that incorporates the additional requirement:

def lrgst(df, cols, sp_col, n):
    for key, col in df.set_index('Col4')[cols].items():
        x = col.nlargest(n)
        J = ', '.join(f'{a}[{b}, {c}]' for a, b, c in zip(x.index, x, df.loc[df['Col4'].isin(x.index), sp_col].values))
        return J

Here's how you can use this updated function:

print(f"{lrgst(df, ['Col1'], 'Col3', 2)}")

This will give you the desired output:

1600[149,309], 1550[148,308]

Let me know if you need any further assistance!