Given project compares various machine learning classifiers and provides their results on car dataset of UCI Machine learning library.
-
Here, python with Jupyter Notebook has been used to implement different classifiers like Decision Tree, Perceptron, Neural Network, Deep Learning, Support Vector Machines, Logistic Regression, Naive Bayes, k Nearest Neighbors, Bagging, Random Forests, Adaboost and Gradient Boosting.
-
For evaluation of classifiers, Car dataset of UCI repository is used. ( https://archive.ics.uci.edu/ml/datasets/Car+Evaluation) Here pandas librery is used to read and preprocess the dataset.
-
Scikit learn package is used to fit and test the model for all classifiers.
-
To run the code, you need to run all cells of MLClassifierComparision.ipynb in Jupyter Notebook.
-
At the beginning we have hyper tuned the code using k-folds. After that we are running all classiifers on the same datasets and testing their accuracy and f_score.