read_csv parameters (sep, header, index_col) in Pandas

Introduction

We use these parameters to tell pandas how to read a CSV file correctly. They help pandas understand the file's structure so it can make a neat table.

When your CSV file uses a different character than a comma to separate values, like a tab or semicolon.

When your CSV file has a header row that names the columns, or when it doesn't.

When you want to use one of the columns as the row labels (index) instead of default numbers.

Syntax

Pandas

pandas.read_csv(filepath, sep=',', header='infer', index_col=None)

sep tells pandas what character separates the columns. Default is a comma.

header tells pandas which row has the column names. Default is the first row (0). Use None if no header.

Examples

Reads a CSV where columns are separated by semicolons instead of commas.

Pandas

pd.read_csv('data.csv', sep=';')

Reads a CSV file that has no header row, so pandas will assign default column names (0, 1, 2, ...).

Pandas

pd.read_csv('data.csv', header=None)

Uses the first column as the row labels (index) instead of default numbers.

Pandas

pd.read_csv('data.csv', index_col=0)

Reads a tab-separated file, uses the first row as header, and sets the column named 'ID' as the index.

Pandas

pd.read_csv('data.csv', sep='\t', header=0, index_col='ID')

Sample Program

This code reads a CSV string with semicolons separating values. It uses the first row as column names and sets the 'ID' column as the index. Then it prints the resulting table.

Pandas

import pandas as pd
from io import StringIO

# Sample CSV data as a string
csv_data = '''ID;Name;Age
1;Alice;30
2;Bob;25
3;Charlie;35'''

# Use StringIO to simulate a file
file_like = StringIO(csv_data)

# Read CSV with semicolon separator, header in first row, and 'ID' as index
df = pd.read_csv(file_like, sep=';', header=0, index_col='ID')

print(df)

Important Notes

If you set header=None, pandas will assign numbers as column names starting from 0.

The index_col can be a column name or a column number (starting at 0).

Using the wrong sep can cause pandas to read the whole line as one column.

Summary

sep tells pandas what separates columns.

header tells pandas which row has column names or if there is none.

index_col sets which column to use as row labels.