Python Profilers: Learn The Basics Of A Profiler For Python

Python Profilers Learn The Basics Of A Profiler For Python

Python profiling tools enable you to understand how fast your code executes. Also, they help you identify the bottlenecks. They play a major role in optimizing your program. It gives you several advantages. For instance, the change in business needs may require you to run code faster. With Python profilers, you can identify the parts of code that are causing sluggish performance. In this article, you will find different ways to profile your Python code. Let’s dive in.

Python Profilers: Learn The Basics Of A Profiler For Python. What are Python profilers?

What are Python profilers?

Python profilers are tools that help you to figure out how the time is spent on the program. It enables you to identify parts that are slowing down the performance. For instance, Application Performance Monitoring (APM) is one of the top profiling tools. You can use it to profile the entire life cycle of transactions for the web application. Hence, you can determine the bottlenecks in your code.

Python Profilers: Learn The Basics Of A Profiler For Python - a close up of a developer with red hair looking at a screenful of code

Python Profilers – How can I profile my Python code?

Python Profilers – Use timeit in Command-Line for Profiling Python Code

Python has a module, called timeit. It enables you to time small code snippets. It utilizes platform-specific time functions. As a result, you will get the most accurate timings.

Let’s take a look at an example. Start the terminal on your PC. Then run the following commands:

Here, you are calling Python on the command line. Then you are passing the “-m” option to look up and use the timeit module as the main program. Next, you pass the “-s” option to tell the module to run the setup once. The command runs the code s 3 times. It returns the best average of the runs.

The -s option is usually used to import the libraries. For instance, you can utilize it to compare the square root function from Python’s math module from NumPy. Here is the code:

Now, let’s write a silly function. You can time it from the command line.

The function will cause an error. However, it is promptly ignored.

Now, you have to run this command:

Here, you are importing and calling the function. Next, you will learn to use timeit inside an actual Python script.

Read: 5 Real-Life Lessons About The Best IDE for Python

Python Profilers – Import timeit for Testing

You can easily use the timeit module inside your code. Here is an example:

Python Profilers – Use a Decorator

You can consider writing your own timer. However, it may not be as accurate as using timeit.

Let’s write the custom decorator:

Here, you are importing the random and time modules from Python’s standard library. Then you are creating your decorator function. It accepts a function. Also, it has another function inside. Before calling the passed-in function, the nested function will grab the time. Next, it waits for the function to return and grabs the end time. Then you will know the time the function took to run. You can print it out. The last two statements return the result of the function call and the function itself.

Python Profilers – Create a Timing Context Manager

You can consider using context managers. It will enable you to time small pieces of code. Let’s create the timer context manager class. Here is the code:

Python Profilers – Using cProfile

Python features built-in code profilers. They are known as the profile module and the cProfile module. The first one is pure Python. It will add a lot of overhead to anything you profile. That’s why you should go with cProfile. It has a similar interface. However, it is significantly faster.

Read: What is the best Python editor?

Let’s look at an example:

What do these columns mean? Let’s find it out.

  • ncalls – It’s the number of calls made.
  • tottime – It refers to the total time spent in the given function.
  • percall – It refers to the quotient of tottime divided by ncalls
  • cumtime – It is the cumulative time spent in this function, along with all subfunctions. For recursive functions, you will find it to be very accurate
  • The second percall column – It refers to the quotient of cumtime divided by primitive calls
  • filename:lineno(function) – It provides the respective data of each function

You can call cProfile on the command line in almost the same way as the timeit module. However, there is a difference. You have to pass a Python script, rather than a snippet. Here’s an example call:

Python Profilers – Use line_profiler

You can use a 3rd party project, called line_profiler. It enables you to profile the time each line takes to execute. The installation process is very simple. You just need to use pip. Here is the command that you need to utilize:

To use line_profiler, you will need some code to profile. You just need to utilize this command:

You will see this output

Now, let’s write the script:

Now, you have to run this command:

You will see this output:

Python Profilers – Use memory_profiler

memory_profiler is an amazing 3rd party profiling package. You can use it to monitor memory consumption in a process. Also, you can use it for the line-by-line analysis of your code.

Read: 6 Best GUI in Python Frameworks

To install memory_profiler, you need to use pip:

Once the installation is done, you have to run a few lines of code against it. Here’s a simple example:

Now, you can run the code by using this command:

Python Profilers – Use profilehooks

Profilehooks is a collection of decorators. It is designed for profiling functions. To install profilehooks, you have to run this command:

Next, you have to write the script:

Now, let’s run the code. You will see this output:

Python Profilers: Learn The Basics Of A Profiler For Python - an overhead view of a laptop with a Python book next to it

Python Profilers – Additional Resources

The profiler of Python’s standard library is very powerful. However, there are plenty of other options to choose from. For example, you can consider using the Python Call Graph module. It visualizes how functions call each other using the GraphViz tool.

Also, you can dig into the compiled code by using Valgrind.

However, you may have to recompile the Python interpreter to enable debugging support.


In this article, you have learned to use timeit and cProfile modules to time and profile your code. Also, you have learned about using the decorator and the context manager. Finally, you explore different 3rd party packages, including line_profiler, memory_profilerand profilehooks. Now, you are ready to use Python profilers. Therefore, you can easily find the bottlenecks in your code.

PyScripter is a powerful Python IDE built on Delphi. It offers several amazing features, including the remote engine, remote debugger, and integrated unit testing. Try it now for free. 


What is the Python standard library?

The Python Standard Library is a collection of script modules accessible to a Python program. It enables you to simplify the programming process. Also, it eliminates the need to rewrite commonly used commands.

How can I optimize the memory usage in my Python program?

You can optimize memory usage using profilers. It enables you to find memory leaks. Hence, you can efficiently optimize your Python programs.

What are function calls?

Function calls are expressions that pass control and arguments to a function.

What is a profiler in programming?

A profiler is a performance analysis tool. It measures the frequency and duration of function calls.

What is debugging in Python?

Debugging is the process of detecting and removing existing and potential errors. It enables you to flawlessly run your Python program.

Related posts

Embrace The Power Of Brand New Python 3.11 Features


How To Exit A Python Script

CodeLearn PythonPythonWindows

How To Add Python Testing Tools Into Pandas Machine Learning Code

Learn PythonPython

7 Easy Steps To Learning Python Scripting

Leave a Reply

Your email address will not be published.