Reading CSV files is one of the most common tasks in data analysis. Here's how to do it efficiently with pandas and the built-in csv module:

1. Basic CSV Reading with Pandas

import pandas as pd

# Read CSV file ✅
df = pd.read_csv('data.csv')

# Display first rows
print(df.head())

# Check shape
print(df.shape)  # (rows, columns)

2. Handle Different Delimiters

# Tab-separated values
df = pd.read_csv('data.tsv', sep='\t')

# Semicolon-separated
df = pd.read_csv('data.csv', sep=';')

# Custom delimiter
df = pd.read_csv('data.txt', sep='|')

# Auto-detect delimiter (slower)
df = pd.read_csv('data.csv', sep=None, engine='python')

3. Specify Column Data Types

# Define data types for better performance
dtypes = {
    'user_id': 'int64',
    'name': 'string',
    'age': 'int64',
    'salary': 'float64',
    'active': 'bool'
}

df = pd.read_csv('data.csv', dtype=dtypes)

# Parse dates during reading (much faster!)
df = pd.read_csv('data.csv',
                 parse_dates=['created_at', 'updated_at'])

4. Handle Missing Values

# Specify missing value indicators
df = pd.read_csv('data.csv',
                 na_values=['NA', 'N/A', 'null', 'NULL', ''])

# Keep default NaN values and add custom ones
df = pd.read_csv('data.csv',
                 keep_default_na=True,
                 na_values=['missing', 'unknown'])

# Don't interpret any values as NaN
df = pd.read_csv('data.csv', keep_default_na=False)

5. Read Large Files Efficiently

# Read in chunks for large files
chunk_size = 10000
chunks = []

for chunk in pd.read_csv('large_file.csv', chunksize=chunk_size):
    # Process each chunk
    processed = chunk[chunk['age'] > 18]
    chunks.append(processed)

df = pd.concat(chunks, ignore_index=True)

# Read only specific columns (faster)
df = pd.read_csv('data.csv', usecols=['name', 'age', 'city'])

# Skip rows
df = pd.read_csv('data.csv', skiprows=5)  # Skip first 5 rows
df = pd.read_csv('data.csv', nrows=1000)  # Read first 1000 rows only

6. Handle Headers and Column Names

# No header in file
df = pd.read_csv('data.csv', header=None)

# Custom column names
df = pd.read_csv('data.csv',
                 names=['col1', 'col2', 'col3'])

# Header on different row
df = pd.read_csv('data.csv', header=2)  # 3rd row is header

# Use first column as index
df = pd.read_csv('data.csv', index_col=0)

7. Using Built-in csv Module

import csv

# Read CSV with csv.reader
with open('data.csv', 'r') as file:
    csv_reader = csv.reader(file)
    header = next(csv_reader)  # Get header

    for row in csv_reader:
        print(row)  # Each row is a list

# Read as dictionaries (easier to work with)
with open('data.csv', 'r') as file:
    csv_reader = csv.DictReader(file)

    for row in csv_reader:
        print(row['name'], row['age'])  # Access by column name

8. Read from URLs

# Read directly from URL
url = 'https://example.com/data.csv'
df = pd.read_csv(url)

# Read compressed files
df = pd.read_csv('data.csv.gz', compression='gzip')
df = pd.read_csv('data.csv.zip', compression='zip')

# Auto-detect compression
df = pd.read_csv('data.csv.gz', compression='infer')

9. Handle Encoding Issues

# Specify encoding (common for international data)
df = pd.read_csv('data.csv', encoding='utf-8')
df = pd.read_csv('data.csv', encoding='latin-1')
df = pd.read_csv('data.csv', encoding='ISO-8859-1')

# Ignore encoding errors
df = pd.read_csv('data.csv', encoding='utf-8', encoding_errors='ignore')

Best Practices

• ✅ Use pandas for data analysis tasks (faster, more features)
• ✅ Use csv module for simple reading/writing without dependencies
• ✅ Specify dtype to reduce memory usage and improve performance
• ✅ Use parse_dates to handle dates during reading (10x faster)
• ✅ Read large files in chunks to avoid memory issues
• ✅ Use usecols to read only needed columns
• ❌ Don't read entire large files into memory if you only need part of it
• ⚠️ Always check encoding if you see strange characters