Kaggle-Ensemble-Guide

A combination of Model Ensembling methods that is extremely useful for increasing accuracy of Kaggle's submission. For more information: http://mlwave.com/kaggle-ensembling-guide/

Installation:

$ pip install -r requirements.txt

Example:

$ python ./src/correlations.py ./samples/method1.csv ./samples/method2.csv
Finding correlation between: ./samples/method1.csv and ./samples/method2.csv
Column to be measured: Label
Pearson's correlation score: 0.67898
Kendall's correlation score: 0.66667
Spearman's correlation score: 0.71053

$ python ./src/kaggle_vote.py "./samples/method*.csv" "./samples/kaggle_vote.csv"
parsing: ./samples/method1.csv
parsing: ./samples/method2.csv
parsing: ./samples/method3.csv
wrote to ./samples/kaggle_vote.csv


$ python ./src/kaggle_vote.py "./samples/_*.csv" "./samples/kaggle_vote_weighted.csv" "weighted"
parsing: ./samples/_w3_method1.csv
Using weight: 3
parsing: ./samples/_w2_method2.csv
Using weight: 2
parsing: ./samples/_w2_method3.csv
Using weight: 2
wrote to ./samples/kaggle_vote_weighted.csv

$ python ./src/kaggle_rankavg.py "./samples/method*.csv" "./samples/kaggle_rankavg.csv"
parsing: ./samples/method1.csv
parsing: ./samples/method2.csv
parsing: ./samples/method3.csv
wrote to ./samples/kaggle_rankavg.csv

$ python ./src/kaggle_avg.py "./samples/method*.csv" "./samples/kaggle_avg.csv"
parsing: ./samples/method1.csv
parsing: ./samples/method2.csv
parsing: ./samples/method3.csv
wrote to ./samples/kaggle_avg.csv

$ python ./src/kaggle_geomean.py  "./samples/method*.csv" "./samples/kaggle_geomean.csv"
parsing: ./samples/method1.csv
parsing: ./samples/method2.csv
parsing: ./samples/method3.csv
wrote to ./samples/kaggle_geomean.csv

Result:

==> ./samples/method1.csv <==
ImageId,Label
1,1
2,0
3,9
4,9
5,3

==> ./samples/method2.csv <==
ImageId,Label
1,2
2,0
3,6
4,2
5,3

==> ./samples/method3.csv <==
ImageId,Label
1,2
2,0
3,9
4,2
5,3

==> ./samples/kaggle_avg.csv <==
ImageId,Label
1,1.666667
2,0.000000
3,8.000000
4,4.333333
5,3.000000

==> ./samples/kaggle_rankavg.csv <==
ImageId,Label
1,0.25
2,0.0
3,1.0
4,0.5
5,0.75

==> ./samples/kaggle_vote.csv <==
ImageId,Label
1,2
2,0
3,9
4,2
5,3

==> ./samples/kaggle_geomean.csv <==
ImageId,Label
1,1.587401
2,0.000000
3,7.862224
4,3.301927
5,3.000000