Contents of this repository :
-
Decision Tree Implementation: Implement your own version of the decision tree using binary univariate split, entropy and information gain. ( WITHOUT using Scikit Learn/sklearn or other libraries)
-
Experimenting with different Kernels ( linear, polynomial, RBF etc) and different levels of regularalization for SVM using sklearn
-
Random Forest Implementation: Implement your own version of the decision tree using binary univariate split, entropy and information gain. ( WITHOUT using Scikit Learn/sklearn or other libraries)
-
Training a Gradient Boost Classifier on tabular loan data post suitable preprocessing ( missing value imputation, categorical variable handling etc)
-
ML Hackathon : Missing Value Imputation, Column wise EDA , feature engineering ( e.g. crafting various time features from the Crash_date_time) & finally training the CATBOOST model.