5 Quick Ways to Add a Max Column in Pandas Without Loops

Introduction

A fan asked how to create a new column in a Pandas DataFrame that contains the row‑wise maximum of two existing columns. The original loop‑based approach works but is cumbersome.

Solution Process

Five different methods are provided, each demonstrating a clean way to compute the max column.

Method 1: Direct max()

Use Pandas' built‑in max function across the two columns.

df['max1'] = df[['cell1','cell2']].max(axis=1)
df

Method 2: loc with max()

Apply max via .loc to achieve the same result.

df['max2'] = df.loc[:,['cell1','cell2']].max(axis=1)
df

Method 3: apply(max, axis=1)

Leverage the apply function with Python's built‑in max.

df['max3'] = df[['cell1','cell2']].apply(max, axis=1)
df

Method 4: assign(new=...)

Use assign to create the new column directly.

df = df.assign(new=df[['cell1','cell2']].max(1))

Note the subtle syntax details to avoid errors.

Method 5: NumPy where()

Combine NumPy's where with Pandas for a vectorised solution.

df['max4'] = np.where(df['cell1'] > df['cell2'], df['cell1'], df['cell2'])
df

Conclusion

The five approaches demonstrate how to efficiently add a row‑wise maximum column in Pandas, offering alternatives that avoid explicit loops and cater to different coding preferences.