This repository consists of the work done for the course "Introduction to Machine Learning and Pattern Recognition".
erm_classification.py: This code simulates a binary classification problem with multivariate Gaussian classes, computes empirical and theoretical decision thresholds, and plots the resulting ROC curve with minimum error points highlighted.
gaussian_mixture.py: This code generates and classifies synthetic 3D data from a mixture of Gaussians using a Bayes classifier with varying loss matrices, then visualizes the true and predicted class distributions.
human_acitivity.py: This code implements a Bayesian classifier to predict human activity classes from feature data using class-specific Gaussian distributions, evaluates its performance, and visualizes the results using PCA.
wine.py: This code implements a Bayesian classifier on the white wine quality dataset using class-conditional Gaussian models, evaluates performance with a confusion matrix and error rate, and visualizes the data in 3D using PCA.
model.py: This code forecasts monthly item sales using Linear Regression, XGBoost, and Random Forest models, and compares their performance with actual sales.