Pet-Adoption---HackerEarth-Hackathon

My Machine Learning Model got the accuracy of 87.38717% in the "HackerEarth Machine Learning challenge: Pet Adoption Hackathon" and my position was 401st among 24375 accepted submissions.

Methodology

[1] MLPClassifier

[2] KNeighborsClassifier

[3] SVC

[4] GaussianProcessClassifier

[5] RBF

[6] DecisionTreeClassifier

[7] RandomForestClassifier

[8] AdaBoostClassifier

[9] GaussianNB

[10] QuadraticDiscriminantAnalysis

[11] SGDClassifier

These classifiers have been used as ML models to classify both the 'breed_category' and 'pet_category' as target variables. The features were condition, color, length (in meter), height (in cm), X1, X2 (X1, X2 are anonymous columns) of the pet. Among all the ML models I've used so far, RandomForest classifier performed the best. When I took part in this competition, I was absolutely new to Python, Data Science, and Machine Learning domain. Hence, it was a wonderful start and score for me utilizing familiar approaches and methodologies in this hackathon.

Thanks to Akshata Kanumuri (Quantitative Operations Assoc at Bank of America) who inspired me and mentored me throughout this whole journey of Data Science and Machine learning hackathons.

Leaderboard (Posotion 401st)

HackerEarth Machine Learning Challenge: Pet Adoption Leaderboard: Position 401st