How to Count Specific Characters in a Pandas Series (Case‑Insensitive)

Rescue Pandas Series: Counting Character Occurrences

Many users avoid pandas for data manipulation, so this article shows how to make pandas more appealing by counting how often specific characters appear in a Series.

Platform

Windows 10

Python 3.8

pandas >= 1.2.4

Data Requirement

Construct a test Series and count the number of times a given character appears, ignoring case.

import pandas as pd

s = pd.Series(['Brown', 'Golden', 'Oracle', 'Mysql', 'Python', 'White', 'Apple'])

Requirement Processing

Count how many times the character 'p' appears in each string, case‑insensitive.

Method 1: List Comprehension

s.map(lambda x: len([i for i in x.lower() if i == 'p']))

Convert each string to lower case and count 'p' occurrences.

Method 2: str.count

Use the pandas str.count method after converting to lower case.

s.str.lower().str.count('p')

Both methods produce the same result.

Method 3: str.findall for Multiple Characters

To count two characters ('a' and 'p'), extend the condition or use a regular expression.

s.str.findall(r'[ap]', flags=re.I).str.len()

Method 4: str.extractall

Similar to findall, extractall returns a DataFrame with a multi‑index.

s.str.extractall(r'([ap])', flags=re.I).groupby(level=0).size()

The result can be aggregated with groupby.filter if needed.

Method 5: extractall with Grouping for Complex Conditions

Count strings that contain at least one character from each of two groups (e.g., 'ap' and 'oe').

s.str.extractall(r'([ap])?([oe])?', flags=re.I).groupby(level=0).count().replace({0: pd.NA}).sum(axis=1, skipna=False)

Summary

The article walks through counting single and multiple characters in a pandas Series, showcasing list comprehensions, str.count, str.findall, and str.extractall. Each method has trade‑offs in readability and scalability, offering flexible solutions for various string‑matching needs.