How to Measure Elapsed Time in Python

Method 2: Using the time.perf_counter() function in Python

The time.perf_counter() function in Python provides a higher resolution timer that doesn’t change even when the system clock changes, making it ideal for benchmarking.

Example:

Output:

Explanation: Here, this function in Python performs much better for smaller durations as it gives a high-resolution timer that remains unaffected by system clock changes or updates.

Pros of time.perf_counter() in Python

• The method carries high resolution.
• Not affected by system clock changes.
• Best for short code and testing for performance.

Cons of time.perf_counter() in Python

• Only available in versions above Python 3.3.
• This method is also not consistent with the total sampling time and may vary each time.
• Not exactly human-readable as it returns a float.

Method 3: Using the time.process_time() function in Python 

time.process_time() in Python measures the time registered in the CPU taken by the program, excluding the sleep time.

Example:

Python

Copy Code

Code Copied!

Close Code

Output:

Explanation: Here, the function is useful for measuring CPU-intensive tasks. It also ignores time spent in sleep or waiting for input/output.

Pros of time.process_time() in Python

• Measures the actual time taken by the CPU.
• This method also ignores I/O wait time and sleep.
• Highly consistent in CPU measurements.

Cons of time.process_time() in Python

• Not the best method for measuring total execution time.
• Not the best method to rely on as it cannot give precise time.
• Not useful in the case of wall-clock time benchmarking.

Get 100% Hike!

Master Most in Demand Skills Now!

By providing your contact details, you agree to our Terms of Use & Privacy Policy

Method 4: Using the datetime Module for Elapsed Time in Python

This particular module in Python gives us the datetime.now() for the measurement of elapsed time.

Example:

Python

Copy Code

Code Copied!

Close Code

Output:

Explanation: Here, the date.timedelta object is responsible for giving the elapsed time in a human-readable format.

Pros of datetime.now() in Python

• This method has a human-readable format.
• This method is useful for logging timestamps.
• Can provide full-time stamps with time and date.

Cons of datetime.now() in Python

• Lower precision compared to time.per_counter()
• Lower accuracy as well, thus not a trustworthy output. Also, garbage collection can occur.
• Can get affected by changes in the system clock.

Method 5: Using the timeit Module in Python 

This module in Python provides a better and accurate way to measure execution time for small code snippets.

Example:

Python

Copy Code

Code Copied!

Close Code

Output:

Explanation: The timeit.timeit() runs the function multiple times, in this case 1500 times, and then reports the total time taken. This makes the method more reliable for its performance benchmarking.

Pros of timeit Module in Python

• It’s best suited for benchmarking.
• It also eliminates overhead from other operations.
• Also eliminates overhead from unrelated operations.

Cons of timeit Module in Python

• This method is not suitable in the long run.
• Also, the results carry variability.
• A little harder to use for complex and parameterized functions.

Comparison of Elapsed Time Measurement Methods in Python

Method	Precision	Uses System Clock	When to Use It
time.time()	Low as it depends on the system clock, which is not very precise	Yes	General time measurement
time.perf_counter()	High as it uses a steady timer, which is also good for small time gaps	No	Precise benchmarking
time.process_time()	High as it counts only CPU time and skips waiting time	No	CPU execution time
datetime.now()	Medium as it uses the system clock, which is better for general use	Yes	Human-readable logs
timeit.timeit()	Very high as it runs code many times, which helps in filtering out the noise	No	Code benchmarking

Real-World Applications of Measuring Elapsed Time in Python

In the segment below, you will learn to use measuring techniques for elapsed time in the real world.

Example 1: Measuring response time for API implementation time.perf_counter()

Python

Copy Code

Code Copied!

Close Code

Output

Explanation:  Here, initially, the timer starts using time.perf_counter(), and then we make a GET request for a sample API. Then, when the response is parsed and the requests for completion, the difference will give us how much time the API call took.

Example 2: Using timeit to benchmark 2 data cleaning methods

Python

Copy Code

Code Copied!

Close Code

Output

Explanation: Here, the timeit.timeit() function is being used to run both the cleaning functions 1000 times. You will have to pass the function name as a string and import it using the setup parameter.

Best Practices to Measure Elapsed Time in Python

• Whenever you decide to measure elapsed time in coding, choose the best method based on your precision requirements.
• For basic benchmarking of code snippets, you should use the timeit module.
• Keep in mind to avoid using time.time() for short-duration measurements.
• Ensure external factors like background processes do not impact results.
• Run the measurement multiple times to get a more reliable average.

Unlock Your Future in Python

Start your Python Journey for Free Today

Explore Program

Conclusion

Elapsed time in Python can be measured using different methods depending on what you need. For basic timing, time.time() is good, while time.perf_counter() gives more accurate results. If you want to measure CPU usage, use time.process_time(). For testing how long specific code takes, timeit.timeit() is helpful. Python has many tools to help you check and improve performance based on your needs.

Further, check out our Python certification course and get ready to excel in your career with our Basic Python interview questions prepared by experts.

FAQs – How to Measure Elapsed Time in Python

Q1. Can time.time() measure execution time accurately?

The time.time() method is a suitable method for general time measurement, but it also lacks precision for short durations.

Q2. Situation where I can use time.perf_counter() over time.time()?

A situation where you would need high resolution, precise timing is not affected by system clock changes.

Q3. Can I change the unit of measured time to milliseconds?

Yes, you can multiply the elapsed time by 1000 units to get milliseconds by writing 

‘elapsed_time * 1000’.

Q4. Best possible method to benchmark a Python function?

You would need to minimize external influence and provide accurate results by using the timeit module.

Q5 How does timeit handle multiple runs for accuracy?

The module runs the function multiple times and averages the execution time to reduce the variability.