FunMark

A package to bench-mark python fun-ctions for runtime and memory usage and to using methods like plotting to view and compare the results easy!

Checkout the production branch for files used to create pypi package

Installation

pip install funmark

Right now the library isn't uploaded to pypi so this won't work as of now

QuickStart

Importing Benchmarking module of funmark

Funmark has class name Benchmark which you can import using

import funmark
# Bench is an object of Benchmark class
bench = funmark.Benchmark()

or

from funmark import Benchmark
# Bench is an object of Benchmark class
bench = Benchmark()

Calculating Runtime and memory consumed by a function

def doSomecCalculation(aList, i, j):
    '''
    doing some calculations
    '''
    return aList

bench = Benchmark()

'''
Benchmark.run() runs your function and returns time and memory used up by your function in seconds and KB
logs: True/Fasle(True by default)(not necessary) Prints information about runtime and memory after every run statment
'''

time, memory = bench.run(doSomecCalculation, aList, i, j, logs=True)

'''
Benchmark.add() saves the record to its object
The first parameter is the X-cordinate while drawing the graph, it can be anything like length of list or size of graph or equivalent
'''

bench.add(len(aList), time, memory)

Make Sure you do add many points with different values to get a good result while representing your data in a graph

Plotting Time and Memory used by function in Graph

'''
Input:
xlabel: str, Title of X-axis
ylabel: str, Title of Y-axis
title: Title of plot
label: Label for legend in graph
show: True/False, Used to toggle the last line plt.show() | Ture -> It will plot the graph

Output:
plotObject: matplotlib.pyplot Object, you can use this object to further make changes in the plot just make sure to set show=False, If the graph is plot then you won't be able make any channges
'''
plotObject = bench.plotTime(
        xlabel="Size of array",
        ylabel="Time Taken",
        title="Time vs size of list",
        label="doSomecCalculation function",
        show=True
                )

'''
Input:
xlabel: str, Title of X-axis
ylabel: str, Title of Y-axis
title: Title of plot
label: Label for legend in graph
show: True/False, Used to toggle the last line plt.show() | Ture -> It will plot the graph

Output:
plotObject: matplotlib.pyplot Object, you can use this object to further make changes in the plot just make sure to set show=False, If the graph is plot then you won't be able make any channges
'''

plotObject = bench.plotMemory(
        xlabel="Size of array",
        ylabel="Time Taken",
        title="Time vs size of list",
        label="doSomecCalculation function",
        show=True
                )

Comparing performance of two different functions

'''
Input:
xlabel: str, Title of X-axis
ylabel: str, Title of Y-axis
title: Title of plot

Output:
A Graph with combined plot of all Benchmark Objects passed in parameter including the one from which method is called.
'''

bench_1.compareTime(
        xlabel="Size of array",
        ylabel="Time Taken",
        title="Time vs size of list",
        bench_2,
        bench_3,
        .
        .
        bench_n
)

bench_1.compareMemory(
        xlabel="Size of array",
        ylabel="Time Taken",
        title="Time vs size of list",
        bench_2,
        bench_3,
        .
        .
        bench_n
)