Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Pull Request for the Algorithm Comparison Tool #4

Open
wants to merge 8 commits into
base: master
Choose a base branch
from
Open

Pull Request for the Algorithm Comparison Tool #4

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
9a4f2da
Knowledge Linker Algorithm Comparison tool, taking the output files f…
marvram Jan 24, 2017
358ce22
Fixed and added some functionality: removed getopt import, used argpa…
Mar 15, 2017
36f2d7b
Update README.md
mihaivavram Mar 15, 2017
eb8c46f
Removed output directory and tool directory, kept input files as exam…
Mar 15, 2017
669002a
Merge branch 'master' of https://github.com/mihaivavram/knowledge_linker
Mar 15, 2017
7abda19
AlgorithmComparisonTool: Changing input format directive and creating…
mihaivavram Jun 27, 2017
a5f3b3b
Update README.md
mihaivavram Jun 27, 2017
1e5a61e
Update README.md
mihaivavram Jun 27, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
321 changes: 321 additions & 0 deletions scripts/algcomptool/AlgCompTool.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,321 @@
#Author: Mihai Avram - [email protected]
#Date: 06/27/2017

import sys
import argparse
import pandas as pd
import numpy as np
import string

def main():
parser = argparse.ArgumentParser()
#Required parameters inputFile1 and inputFile2 which denote the paths to the two input algorithm files to be compared (print output to the console)
parser.add_argument("inputFile1", type=str, help="the relative or absolute location and name of the first input file")
parser.add_argument("inputFile2", type=str, help="the relative or absolute location and name of the second input file")
#Optional parameters -oc, -o1, -o2 which denote output paths for the comparisons (print output to these files)
parser.add_argument("-oc", "--outputcomparison", type=str, help="the relative or absolute location and name of the file 1 vs. file 2 comparison output file (.txt extension recommended)", required=False, default="CompOutputDefault.txt")
parser.add_argument("-o1", "--file1Confusion", type=str, help="the relative or absolute location and name of the first confusion result output file (.csv extension recommended)", required=False, default="File1OutConfDefault.csv")
parser.add_argument("-o2", "--file2Confusion", type=str, help="the relative or absolute location and name of the second confusion result output file (.csv extension recommended)", required=False, default="File2OutConfDefault.csv")
args = parser.parse_args()

computeAndPrint(args.inputFile1, args.inputFile2, args.outputcomparison, args.file1Confusion, args.file2Confusion)

def computeAndPrint(inputFile1, inputFile2, outputcomparison, file1Confusion, file2Confusion):
#Importing algorithms to be compared
try:
alg1 = pd.read_csv(inputFile1)
alg2 = pd.read_csv(inputFile2)
except:
print("Error reading the alg1 and alg2 input files, please ensure file path and file format is correct.", file=sys.stderr)
sys.exit(1)

#Alg1 checking for null values in pertinent columns
if any(pd.isnull(alg1['sid'])) or any(pd.isnull(alg1['oid'])) or any(pd.isnull(alg1['class'])) or any(pd.isnull(alg1['score'])):
print("There are missing column values in the sid, oid, class, or score columns in alg1", file=sys.stderr)
sys.exit(1)

#Alg2 checking for null values in pertinent columns
if any(pd.isnull(alg2['sid'])) or any(pd.isnull(alg2['oid'])) or any(pd.isnull(alg2['class'])) or any(pd.isnull(alg2['score'])):
print("There are missing column values in the sid, oid, class, or score columns in alg2", file=sys.stderr)
sys.exit(1)

#Algorithm comparison
#Preparatory work
alg1 = alg1.sort_values(['sid','score'], axis=0, ascending=False)
alg2 = alg2.sort_values(['sid','score'], axis=0, ascending=False)

alg1DistinctObjNum = len(alg1.oid.unique())
alg2DistinctObjNum = len(alg2.oid.unique())

idColLocation = alg1.columns.get_loc('sid')
subjectColLocation = alg1.columns.get_loc('sub')
objectIDColLocation = alg1.columns.get_loc('oid')
objectColLocation = alg1.columns.get_loc('obj')
classColLocation = alg1.columns.get_loc('class')
scoreColLocation = alg1.columns.get_loc('score')

correctPredAlg1 = {}
incorrectPredAlg1 = {}

correctPredAlg2 = {}
incorrectPredAlg2 = {}

#Creating Dictionary of id/subject matches
idSubjectMapping = {}
for index, row in alg1.iterrows():
idSubjectMapping[row[subjectColLocation]] = row[idColLocation]

#Comparisons
for sample in range(0, alg1.shape[0], alg1DistinctObjNum):
if alg1.iloc[sample, classColLocation] == 1:
correctPredAlg1[str(alg1.iloc[sample, subjectColLocation])] = alg1.iloc[sample, scoreColLocation]
elif alg1.iloc[sample, classColLocation] == 0:
incorrectPredAlg1[str(alg1.iloc[sample, subjectColLocation])] = alg1.iloc[sample, scoreColLocation]


for sample in range(0, alg2.shape[0], alg2DistinctObjNum):
if alg2.iloc[sample, classColLocation] == 1:
correctPredAlg2[str(alg2.iloc[sample, subjectColLocation])] = alg2.iloc[sample, scoreColLocation]
elif alg2.iloc[sample, classColLocation] == 0:
incorrectPredAlg2[str(alg2.iloc[sample, subjectColLocation])] = alg2.iloc[sample, scoreColLocation]


#if the 3 output files were not specified, print results to console, if only one was specified, use that one and default locations for the others
if outputcomparison == "CompOutputDefault.txt" and file1Confusion == "File1OutConfDefault.csv" and file2Confusion == "File2OutConfDefault.csv":
#Printing Results (comparisonPrint function used for this)
def comparisonPrint(algorithm):
for subject, score in algorithm.items():
print("ID:", idSubjectMapping[subject], "- Subject:", subject.replace("dbr:",""), "with score", score)

print("\nALG1 CORRECTLY PREDICTED THE FOLLOWING")
comparisonPrint(correctPredAlg1)

print("\nALG1 INCORRECTLY PREDICTED THE FOLLOWING")
comparisonPrint(incorrectPredAlg1)

print("\nALG2 CORRECTLY PREDICTED THE FOLLOWING")
comparisonPrint(correctPredAlg2)

print("\nALG2 INCORRECTLY PREDICTED THE FOLLOWING")
comparisonPrint(incorrectPredAlg2)

bothCorrect = set(correctPredAlg1.keys()) & set(correctPredAlg2.keys())
bothIncorrect = set(incorrectPredAlg1.keys()) & set(incorrectPredAlg2.keys())

alg1Correctalg2Incorrect = set(correctPredAlg1.keys()) & set(incorrectPredAlg2.keys())
alg2Correctalg1Incorrect = set(correctPredAlg2.keys()) & set(incorrectPredAlg1.keys())

print("\nINSTANCES WHERE ALG1 PREDICTED CORRECTLY BUT ALG2 DIDN'T")
for instance in alg1Correctalg2Incorrect:
print("ID:",idSubjectMapping[instance], "- Subject:", instance.replace("dbr:",""))

print("\nINSTANCES WHERE ALG2 PREDICTED CORRECTLY BUT ALG1 DIDN'T")
for instance in alg2Correctalg1Incorrect:
print("ID:",idSubjectMapping[instance], "- Subject:", instance.replace("dbr:",""))

#Comparison of algorithms for when both predicted correctly
print("\nCOMPARISON OF ALGS WHEN BOTH PREDICTED CORRECTLY")
for subject in bothCorrect:

alg1CompSegment = alg1[alg1['sub'] == subject]
alg2CompSegment = alg2[alg2['sub'] == subject]

#Same prediction level, so must compare ratios between our first correct prediction, and the next prediction
alg1Ratio = alg1CompSegment.iloc[0,scoreColLocation]/alg1CompSegment.iloc[1,scoreColLocation]
alg2Ratio = alg2CompSegment.iloc[0,scoreColLocation]/alg2CompSegment.iloc[1,scoreColLocation]

#The greater the ratio the better the prediction, i.e.
#0.5 (our prediction) 0.04 (next prediction) vs.
#0.5 (our prediction) 0.4 (next prediction)
if alg1Ratio > alg2Ratio:
print("Alg1 > Alg2 for relation:", "ID:", idSubjectMapping[subject], "- Subject:", subject.replace("dbr:",""))
elif alg1Ratio < alg2Ratio:
print("Alg1 < Alg2 for relation:", "ID:", idSubjectMapping[subject], "- Subject:", subject.replace("dbr:",""))



print("\nCOMPARISON OF ALGS WHEN BOTH PREDICTED INCORRECTLY")
#Comparison of algorithms for when both predicted incorrectly
for subject in bothIncorrect:
#Alg1 Analysis
alg1CompSegment = alg1[alg1['sub'] == subject]
alg1Sample = 0
while alg1CompSegment.iloc[alg1Sample,classColLocation] != 1:
alg1Sample = alg1Sample + 1

#Alg2 Analysis
alg2CompSegment = alg2[alg2['sub'] == subject]
alg2Sample = 0
while alg2CompSegment.iloc[alg2Sample,classColLocation] != 1:
alg2Sample = alg2Sample + 1

if alg1Sample > alg2Sample:
print("Alg1 < Alg2 for relation:", "ID:", idSubjectMapping[subject], "- Subject:", subject.repalce("dbr:",""))
elif alg1Sample < alg2Sample:
print("Alg1 > Alg2 for relation:", "ID:", idSubjectMapping[subject], "- Subject:", subject.replace("dbr:",""))
elif alg1Sample == alg2Sample:
#Same prediction level, so must compare ratios between highest weight prediction and this wrong preiction
alg1Ratio = alg1CompSegment.iloc[0,scoreColLocation]/alg1CompSegment.iloc[alg1Sample,scoreColLocation]
alg2Ratio = alg2CompSegment.iloc[0,scoreColLocation]/alg2CompSegment.iloc[alg2Sample,scoreColLocation]
#The bigger the ratio the worse the prediction: i.e. 0.5 (first), 0.3, 0.001, 0.0001 (our prediction) vs.
#0.5 (first), 0.2 (our prediction)
if alg1Ratio > alg2Ratio:
print("Alg1 < Alg2 for relation:", "ID:", idSubjectMapping[subject], "- Subject:", subject.repalce("dbr:",""))
elif alg1Ratio < alg2Ratio:
print("Alg1 > Alg2 for relation:", "ID:", idSubjectMapping[subject], "- Subject:", subject.replace("dbr:",""))



print("\n\nALGORITHM 1: CALCULATING CONFUSION METRICS I.E. TRUE POSITIVE, FALSE POSITIVE, ETC...\n\n")

for sample in range(0, alg1.shape[0]):
if sample % alg1DistinctObjNum == 0:
if alg1.iloc[sample, classColLocation] == 1:
print("SID:", alg1.iloc[sample, idColLocation], "- SUBJECT:", alg1.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg1.iloc[sample, objectIDColLocation], "- OBJECT:", alg1.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "TP")
else:
print("SID:", alg1.iloc[sample, idColLocation], "- SUBJECT:", alg1.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg1.iloc[sample, objectIDColLocation], "- OBJECT:", alg1.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "FP")
else:
if alg1.iloc[sample, classColLocation] == 1:
print("SID:", alg1.iloc[sample, idColLocation], "- SUBJECT:", alg1.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg1.iloc[sample, objectIDColLocation], "- OBJECT:", alg1.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "FN")
else:
print("SID:", alg1.iloc[sample, idColLocation], "- SUBJECT:", alg1.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg1.iloc[sample, objectIDColLocation], "- OBJECT:", alg1.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "TN")

print("\n\nALGORITHM 2: CALCULATING CONFUSION METRICS I.E. TRUE POSITIVE, FALSE POSITIVE, ETC...\n\n")

for sample in range(0, alg2.shape[0]):
if sample % alg2DistinctObjNum == 0:
if alg2.iloc[sample, classColLocation] == 1:
print("SID:", alg2.iloc[sample, idColLocation], "- SUBJECT:", alg2.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg2.iloc[sample, objectIDColLocation], "- OBJECT:", alg2.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "TP")
else:
print("SID:", alg2.iloc[sample, idColLocation], "- SUBJECT:", alg2.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg2.iloc[sample, objectIDColLocation], "- OBJECT:", alg2.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "FP")
else:
if alg2.iloc[sample, classColLocation] == 1:
print("SID:", alg2.iloc[sample, idColLocation], "- SUBJECT:", alg2.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg2.iloc[sample, objectIDColLocation], "- OBJECT:", alg2.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "FN")
else:
print("SID:", alg2.iloc[sample, idColLocation], "- SUBJECT:", alg2.iloc[sample, subjectColLocation].replace("dbr:",""), "- OID:", alg2.iloc[sample, objectIDColLocation], "- OBJECT:", alg2.iloc[sample, objectColLocation].replace("dbr:",""), "- CONFUSION:", "TN")
print("Done.")

#WRITING THE RESULTS TO FILES
else:
try:
fileComp = open(outputcomparison,'w')
except:
print("Error writing to the specified outputcomparison file, please check -oc argument", file=sys.stderr)
sys.exit(1)

#Printing Results (comparisonWrite function used for this)
def comparisonWrite(algorithm):
for subject, score in algorithm.items():
fileComp.write("ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + " with score " + str(score) + "\n")

fileComp.write("\nALG1 CORRECTLY PREDICTED THE FOLLOWING\n")
comparisonWrite(correctPredAlg1)

fileComp.write("\nALG1 INCORRECTLY PREDICTED THE FOLLOWING\n")
comparisonWrite(incorrectPredAlg1)

fileComp.write("\nALG2 CORRECTLY PREDICTED THE FOLLOWING\n")
comparisonWrite(correctPredAlg2)

fileComp.write("\nALG2 INCORRECTLY PREDICTED THE FOLLOWING\n")
comparisonWrite(incorrectPredAlg2)

bothCorrect = set(correctPredAlg1.keys()) & set(correctPredAlg2.keys())
bothIncorrect = set(incorrectPredAlg1.keys()) & set(incorrectPredAlg2.keys())

alg1Correctalg2Incorrect = set(correctPredAlg1.keys()) & set(incorrectPredAlg2.keys())
alg2Correctalg1Incorrect = set(correctPredAlg2.keys()) & set(incorrectPredAlg1.keys())

fileComp.write("\nINSTANCES WHERE ALG1 PREDICTED CORRECTLY BUT ALG2 DIDN'T\n")
for instance in alg1Correctalg2Incorrect:
fileComp.write("ID: " + str(idSubjectMapping[instance]) + " - Subject: " + str(instance.replace("dbr:","")) + "\n")

fileComp.write("\nINSTANCES WHERE ALG2 PREDICTED CORRECTLY BUT ALG1 DIDN'T\n")
for instance in alg2Correctalg1Incorrect:
fileComp.write("ID: " + str(idSubjectMapping[instance]) + " - Subject: " + str(instance.replace("dbr:","")) + "\n")

#Comparison of algorithms for when both predicted correctly
fileComp.write("\nCOMPARISON OF ALGS WHEN BOTH PREDICTED CORRECTLY\n")
for subject in bothCorrect:

alg1CompSegment = alg1[alg1['sub'] == subject]
alg2CompSegment = alg2[alg2['sub'] == subject]

#Same prediction level, so must compare ratios between our first correct prediction, and the next prediction
alg1Ratio = alg1CompSegment.iloc[0,scoreColLocation]/alg1CompSegment.iloc[1,scoreColLocation]
alg2Ratio = alg2CompSegment.iloc[0,scoreColLocation]/alg2CompSegment.iloc[1,scoreColLocation]

#The greater the ratio the better the prediction, i.e.
#0.5 (our prediction) 0.04 (next prediction) vs.
#0.5 (our prediction) 0.4 (next prediction)
if alg1Ratio > alg2Ratio:
fileComp.write("Alg1 > Alg2 for relation: " + " ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + "\n")
elif alg1Ratio < alg2Ratio:
fileComp.write("Alg1 < Alg2 for relation: " + " ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + "\n")

fileComp.write("\nCOMPARISON OF ALGS WHEN BOTH PREDICTED INCORRECTLY\n")
#Comparison of algorithms for when both predicted incorrectly
for subject in bothIncorrect:

#Alg1 Analysis
alg1CompSegment = alg1[alg1['sub'] == subject]
alg1Sample = 0
while alg1CompSegment.iloc[alg1Sample,classColLocation] != 1:
alg1Sample = alg1Sample + 1

#Alg2 Analysis
alg2CompSegment = alg2[alg2['sub'] == subject]
alg2Sample = 0
while alg2CompSegment.iloc[alg2Sample,classColLocation] != 1:
alg2Sample = alg2Sample + 1

if alg1Sample > alg2Sample:
fileComp.write("Alg1 < Alg2 for relation: " + " ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + "\n")
elif alg1Sample < alg2Sample:
fileComp.write("Alg1 > Alg2 for relation: " + " ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + "\n")
elif alg1Sample == alg2Sample:
#Same prediction level, so must compare ratios between highest weight prediction and this wrong preiction
alg1Ratio = alg1CompSegment.iloc[0,scoreColLocation]/alg1CompSegment.iloc[alg1Sample,scoreColLocation]
alg2Ratio = alg2CompSegment.iloc[0,scoreColLocation]/alg2CompSegment.iloc[alg2Sample,scoreColLocation]
#The bigger the ratio the worse the prediction: i.e. 0.5 (first), 0.3, 0.001, 0.0001 (our prediction) vs.
#0.5 (first), 0.2 (our prediction)
if alg1Ratio > alg2Ratio:
fileComp.write("Alg1 < Alg2 for relation: " + " ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + "\n")
elif alg1Ratio < alg2Ratio:
fileComp.write("Alg1 > Alg2 for relation: " + " ID: " + str(idSubjectMapping[subject]) + " - Subject: " + str(subject.replace("dbr:","")) + "\n")
fileComp.close
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Missing function call?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Changing all "missing function calls" you mentioned except this one, because this requires a different output format.


#Function to print confusion result files
def confusionFileWrite(writeFileLoc, algorithm):
try:
fileConfAlg = open(writeFileLoc,"w")
except:
print("Error writing to the specified output file, please check -o1 argument", file=sys.stderr)
sys.exit(1)

fileConfAlg.write("sid,sub,oid,obj,conf\n")

algorithmDistinctObjNum = len(algorithm.oid.unique())

for sample in range(0, algorithm.shape[0]):
if sample % algorithmDistinctObjNum == 0:
if algorithm.iloc[sample, classColLocation] == 1:
fileConfAlg.write(str(algorithm.iloc[sample, idColLocation]) + "," + str(algorithm.iloc[sample, subjectColLocation].replace("dbr:","")) + "," + str(algorithm.iloc[sample, objectIDColLocation]) + "," + str(algorithm.iloc[sample, objectColLocation].replace("dbr:","")) + ",TP" + "\n")
else:
fileConfAlg.write(str(algorithm.iloc[sample, idColLocation]) + "," + str(algorithm.iloc[sample, subjectColLocation].replace("dbr:","")) + "," + str(algorithm.iloc[sample, objectIDColLocation]) + "," + str(algorithm.iloc[sample, objectColLocation].replace("dbr:","")) + ",FP" + "\n")
else:
if algorithm.iloc[sample, classColLocation] == 1:
fileConfAlg.write(str(algorithm.iloc[sample, idColLocation]) + "," + str(algorithm.iloc[sample, subjectColLocation].replace("dbr:","")) + "," + str(algorithm.iloc[sample, objectIDColLocation]) + "," + str(algorithm.iloc[sample, objectColLocation].replace("dbr:","")) + ",FN" + "\n")
else:
fileConfAlg.write(str(algorithm.iloc[sample, idColLocation]) + "," + str(algorithm.iloc[sample, subjectColLocation].replace("dbr:","")) + "," + str(algorithm.iloc[sample, objectIDColLocation]) + "," + str(algorithm.iloc[sample, objectColLocation].replace("dbr:","")) + ",TN" + "\n")

fileConfAlg.close

#Writing confusion result files
confusionFileWrite(file1Confusion, alg1)
confusionFileWrite(file2Confusion, alg2)

print("Done, please check output files.")

if __name__ == "__main__":
main()
39 changes: 39 additions & 0 deletions scripts/algcomptool/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,39 @@
Author: Mihai Avram - [email protected]
</br>
Date: 6/27/2017

## DESCRIPTION:
This tool is used to compare the results of two algorithms (via 2 input files) that have subject to object relations, a score for the given relation, and the class (ground truth) of such relation. The tool compares scores for the two algorithms for the same relations and lists descriptive information about the algorithms, such as which algorithm performed better for which relations and conversely, which algorithm performed worse for which relations. The tool also prints out .csv formatted confusion information (i.e. True Positive, True Negative, False Positive, False Negative) for each algorithm specifically.

## STEPS TO CLEAN DATA:

Note: Look for input/presidentcouplesNODES.csv and input/presidentcouplesRSIM.csv for examples
of what cleaned, proper input files should look like for the input file parameters

1) Ensure the input files have non-null values in titled columns 'sid','score','sub','class','oid','obj'
2) Ensure the input files have (subject)*(object) rows/samples
3) Ensure all values in the 'sub' column for both files have the exact same named entries
(i.e. dbr:A_Beautiful_Mind_(film) in alg1 should be exactly the same name
and not dbresource/:A_Beautiful_Mind_(film) or A_Beautiful_Mind)
4) Ensure the output/ folder is empty as files may get created there and should not
be over-written, if the same three specified output files are listed in multiple iterations of
the tool running, they will be overwritten.

## HOW TO RUN TOOL:
The tool will run in Python 3, and can be executed the following way from the command line:

The following prints the usage of the tool:

`python AlgCompTool.py`

Tool run with required parameters (printing results to the terminal):

`python AlgCompTool.py [INPUTFILE1LOCATION] [INPUTFILE2LOCATION]`

Optional parameters -oc, -o1, and -o2 which are used to place output in output files:

`python AlgCompTool.py [INPUTFILE1LOCATION] [INPUTFILE2LOCATION] -oc [COMPARISONOUTPUTFILELOC] -o1 [OUTPUTFILE1LOCATION] -o2 [OUTPUTFILE2LOCATION]`

## TODO-IMPROVEMENTS:
A possible extension to this script would be to add comparison for multiple algorithms,
not just two. Another possible extension is to parallelize code for bigger data.
Loading