Code Monkey home page Code Monkey logo

execution-timer's Introduction

ExecutionTimer

ExecutionTimer is a utility class for measuring execution times of functions or methods in Python. It provides a decorator to easily measure and optionally save the execution times.

Installation

  • Install via pip:
pip install timer-decorator
  • Install via git clone:
git clone https://github.com/vgilabert94/execution-timer

Documentation

Function: time_execution

Parameters

  • return_measure (bool, optional): Whether to return the measured time along with the function result (default is False).
  • nanoseconds (bool, optional): Whether to use nanoseconds resolution for timing measurements (default is False).
  • n_iter (int, optional): The number of iterations to execute the function/method (default is 1). If n_iter > 1 and return_measure=True: result of the function will be the last execution.
  • return_average (bool, optional): Whether to return the average time when measuring over multiple iterations (default is True). Only is used when n_iter > 1.

Notes

  • If return_measure=False and return_average=False: a message will be printed for each iteration.
  • If n_iter > 1 and return_measure=True: the result will be the last function result.
  • If n_iter > 1 and return_measure=True and return_average=False: the result will be the last function result with a list of times for each iteration.

Class: ExecutionTimer

Parameters

  • save_measure (bool, optional): Flag to determine if measurement results should be saved (default is True).
  • nanoseconds (bool, optional): Flag to use nanoseconds resolution for timing measurements (default is False).
  • n_iter : (int, optional): Number of iterations to execute the function/method (default is 1).

Notes

  • If n_iter > 1 and return_measure=True: the result will be the last function result.
  • If n_iter > 1 and return_measure=False: the time printed will be the last execution result.

Methods

time_execution

Decorator method to measure the execution time of a function or method.

  • func (callable, optional): Function or method to be timed. If None, returns a decorator function.
  • return_measure (bool, optional): Flag to indicate if the measured time should be returned along with the function result.
  • print_measure (bool, optional): Flag to indicate if the measured time should be printed.

Returns:

  • wrapper (callable): Decorated function that measures the execution time of func.

get_measured_time

Retrieve the recorded execution times.

Returns:

  • dict: Dictionary containing the measured execution times. Keys are function or method names, and values are lists of measured times.

reset_measured_time

Reset the recorded execution times.

average_measured_time

Calculate the average execution times for all recorded functions or methods.

Returns:

  • dict: A dictionary where keys are function or class names, and values are the average execution times in seconds. If no times are recorded, the value will be None.

Examples

You can access a comprehensive examples notebook at the following link: examples/notebook.ipynb

Load packages:

import time
from execution_timer import ExecutionTimer, time_execution

Example 1: Measuring a function with the default settings.

timer = ExecutionTimer()

@timer.time_execution
def sample_function(n):
    time.sleep(n)

print(sample_function(n=1))
print(timer.get_measured_time())

Output

None
{'sample_function': [1.0000783540003795]}
@time_execution
def sample_function(n):
    time.sleep(n)

print(sample_function(n=1))

Output

The 'sample_function' function was executed in 1.00085 seconds.
None

Example 2: Measuring N iterations

timer = ExecutionTimer(n_iter=5)

@timer.time_execution
def sample_function(n):
    time.sleep(n)

print(sample_function(n=1))
print(timer.get_measured_time())
print(timer.average_measured_time())

Output

The 'sample_function' function was executed in 1.00098 seconds.
None
{'sample_function': [1.0010039510007118, 1.001013477000015, 1.001078371999938, 1.0008607439995103, 1.0009819289998632]}
{'sample_function': 1.0009876946000076}
@time_execution(n_iter=5)
def sample_function(n):
    time.sleep(n)

print(sample_function(n=1))

Output

The 'sample_function' function was executed in 1.00030 seconds.
None

Example 8: Measuring a Method from a class

timer = ExecutionTimer()

class SampleClass:
    def __init__(self):
        pass

    @timer.time_execution(return_measure=True)
    def sample_method(self, n):
        time.sleep(n)

    @timer.time_execution
    def sample_method_x2(self, n):
        time.sleep(2*n)

sample = SampleClass()
print(sample.sample_method(n=1))
print(sample.sample_method_x2(n=1))
print(timer.get_measured_time())

Output

(None, 1.0002362490004089)
The 'sample_method_x2' function was executed in 2.00203 seconds.
None
{'SampleClass': {'sample_method': [1.0002362490004089], 'sample_method_x2': [2.0020319710001786]}}

LICENSE

Distributed under the MIT License. See LICENSE.txt for more information.

Contact

Vicent Gilabert

Linkedin

execution-timer's People

Contributors

vgilabert94 avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.