How to Get the Number of Rows in a Pandas DataFrame

I’ve spent years wrangling data in Python, and one of the most common things I do is check the size of my dataset.

Whether I am loading a CSV of US Census data or analyzing California housing prices, I always need to know how many records I am dealing with.

Knowing the number of rows helps me verify if my data filters worked or if a merge operation went as expected.

In this tutorial, I will show you the different ways to get the row count of a Pandas DataFrame.

Table of Contents

Use the len() Function

The simplest way to get the number of rows is by using the built-in Python len() function.

I use this method most often because it is readable and very fast. It works exactly like finding the length of a list.

In the example below, I’ll create a DataFrame containing some basic information about major US cities.

import pandas as pd

# Creating a dataset of US Cities
data = {
    'City': ['New York', 'Los Angeles', 'Chicago', 'Houston', 'Phoenix'],
    'State': ['NY', 'CA', 'IL', 'TX', 'AZ'],
    'Population': [8336817, 3979576, 2693976, 2320268, 1680992]
}

df = pd.DataFrame(data)

# Getting the number of rows using len()
row_count = len(df)

print(f"Number of rows: {row_count}")

I executed the above example code and added the screenshot below.

Get the Number of Rows in a Pandas DataFrame

When you pass the DataFrame to len(), it returns the number of rows in the index.

Use the shape Attribute

Another reliable method is using the .shape attribute. I find this particularly useful when I want to know both the number of rows and columns at the same time.

The shape attribute returns a tuple where the first element is the number of rows and the second is the number of columns.

import pandas as pd

# Dataset of US Tech Companies
data = {
    'Company': ['Apple', 'Microsoft', 'Alphabet', 'Amazon', 'Meta'],
    'Headquarters': ['Cupertino', 'Redmond', 'Mountain View', 'Seattle', 'Menlo Park'],
    'Founded': [1976, 1975, 1998, 1994, 2004]
}

df = pd.DataFrame(data)

# Getting dimensions
dimensions = df.shape

# Getting only rows
rows = df.shape[0]

print(f"Dimensions (Rows, Columns): {dimensions}")
print(f"Total Rows: {rows}")

I executed the above example code and added the screenshot below.

How to Get the Number of Rows in a Pandas DataFrame

Since .shape is an attribute and not a function, it doesn’t require parentheses. It is often slightly faster than len() because it accesses the underlying array’s metadata directly.

Use the count() Method

Sometimes, I don’t just want the total number of rows; I want to know how many rows actually contain data.

The .count() method is what I use when I suspect there are missing values (NaNs) in my dataset.

By default, .count() returns the number of non-null values for each column.

import pandas as pd
import numpy as np

# US National Parks dataset with some missing data
data = {
    'Park Name': ['Yellowstone', 'Yosemite', 'Zion', 'Grand Canyon', 'Acadia'],
    'State': ['WY', 'CA', 'UT', 'AZ', np.nan], # Missing state for Acadia
    'Established': [1872, 1890, 1919, 1919, 1919]
}

df = pd.DataFrame(data)

# Using count() to see non-null values
print(df.count())

I executed the above example code and added the screenshot below.

Get the Number of Rows in a Python Pandas DataFrame

In this case, the ‘State’ column will show a count of 4, while the others show 5. This is a lifesaver when cleaning messy US government data files.

Use the len(df.index) Property

If you are working with massive datasets, millions of rows, performance becomes a priority. In my experience, len(df.index) is the fastest way to get the row count.

It skips the overhead of the DataFrame object and goes straight to the index.

import pandas as pd

# Large dataset simulation: US ZIP Codes
# For this example, we'll just use a small representative set
data = {'ZipCode': range(10000, 10010)}
df = pd.DataFrame(data)

# The high-performance way
row_count = len(df.index)

print(f"Fast row count: {row_count}")

While the speed difference is negligible for small files, it adds up quickly in complex data pipelines.

Count Rows Based on a Condition

Often, I need to know how many rows meet a specific criterion.

For example, if I am looking at a list of US States, I might want to know how many have a population over 10 million.

I usually combine a boolean filter with len() or .shape[0] to get this number.

import pandas as pd

# Sample US State Population data (in millions)
data = {
    'State': ['California', 'Texas', 'Florida', 'New York', 'Pennsylvania', 'Illinois'],
    'Population': [39.2, 29.1, 21.5, 20.2, 13.0, 12.8]
}

df = pd.DataFrame(data)

# Count states with population > 20 million
high_pop_count = len(df[df['Population'] > 20])

print(f"States with over 20M residents: {high_pop_count}")

This approach is flexible and allows you to count specific subsets of your data quickly.

Comparison of Methods

Method	What it Returns	Best For
`len(df)`	Total number of rows	General use and readability
`df.shape[0]`	Total number of rows	When you already need columns too
`df.count()`	Non-null values per column	Identifying missing data
`len(df.index)`	Total number of rows	Maximum performance on large data

Which Method Should You Use?

For most of my daily tasks, I stick with len(df). It is clean, Pythonic, and easy for others to read.

If I am building a production-level data application where every millisecond counts, I switch to len(df.index).

And of course, if I’m checking for data quality, df.count() is the only way to go to see those missing values.

I hope this tutorial helped you understand the different ways to find the number of rows in a Pandas DataFrame. Each method has its own place depending on whether you need speed, metadata, or a count of valid data points.

In this guide, I’ve shared the techniques I use every day as a developer. Whether you are dealing with a small list of US cities or a massive database of financial transactions, these tools will help you manage your data more effectively.