Python - Rename column names by index in a Pandas DataFrame without using rename()

Sometimes you need to rename column names in a Pandas DataFrame by their position (index) rather than by name. Python provides a simple way to do this by directly modifying the columns.values array without using the rename() method.

Using columns.values with Index

You can rename columns by accessing their position in the columns.values array. This approach is useful when you know the column positions but not necessarily their current names.

Example

Let's create a DataFrame and rename all columns by their index positions ?

import pandas as pd

# Create DataFrame
dataFrame = pd.DataFrame(
    {
        "Car": ['BMW', 'Lexus', 'Tesla', 'Mustang', 'Mercedes', 'Jaguar'],
        "Reg_Price": [7000, 1500, 5000, 8000, 9000, 6000],
        "Units": [90, 120, 100, 150, 200, 130]
    }
)

print("Original DataFrame:")
print(dataFrame)

# Renaming columns by index
dataFrame.columns.values[0] = "Car Names"
dataFrame.columns.values[1] = "Registration Cost"
dataFrame.columns.values[2] = "Units_Sold"

print("\nDataFrame after renaming columns:")
print(dataFrame)

Original DataFrame:
       Car  Reg_Price  Units
0      BMW       7000     90
1    Lexus       1500    120
2    Tesla       5000    100
3  Mustang       8000    150
4 Mercedes       9000    200
5   Jaguar       6000    130

DataFrame after renaming columns:
  Car Names  Registration Cost  Units_Sold
0       BMW               7000          90
1     Lexus               1500         120
2     Tesla               5000         100
3   Mustang               8000         150
4  Mercedes               9000         200
5    Jaguar               6000         130

Renaming Specific Columns

You can also rename only specific columns by targeting their index positions ?

import pandas as pd

# Create DataFrame
df = pd.DataFrame({
    "A": [1, 2, 3],
    "B": [4, 5, 6], 
    "C": [7, 8, 9]
})

print("Before renaming:")
print(df.columns.tolist())

# Rename only first and last columns
df.columns.values[0] = "First_Column"
df.columns.values[2] = "Last_Column"

print("\nAfter renaming:")
print(df.columns.tolist())
print("\nDataFrame:")
print(df)

Before renaming:
['A', 'B', 'C']

After renaming:
['First_Column', 'B', 'Last_Column']

DataFrame:
   First_Column  B  Last_Column
0             1  4            7
1             2  5            8
2             3  6            9

How It Works

The columns.values property returns a NumPy array containing column names. By accessing specific indices like columns.values[0], you can directly modify individual column names. This method modifies the DataFrame in-place without creating a copy.

Key Points

• Index 0 refers to the first column, index 1 to the second column, and so on
• This method modifies the original DataFrame (in-place operation)
• Negative indices work too: columns.values[-1] refers to the last column
• More memory-efficient than rename() as it doesn't create copies

Conclusion

Using columns.values[index] provides a direct way to rename DataFrame columns by position. This approach is efficient for renaming multiple columns when you know their positions, especially useful in data preprocessing pipelines.