Machine-Learning-fundamentals

Machine learning projects in python with source code

Linear Regression

1. Preprocess the data for linear regression.
2. Compute the cost and perform gradient descent in pure numpy in vectorized form.
3. Fit a linear regression model using a single feature.
4. Visualize my results using matplotlib.
5. Perform multivariate linear regression.
6. Pick the best features in the dataset.
7. Experiment with adaptive learning rates.

Decision Trees

1. Implement two impurity measures: Gini and Entropy.
2. Construct a decision tree algorithm.
3. Prune the tree to achieve better results.
4. Visualize ny results.

MAP Classifier

1. Implement a Naive Bayes Classifier based on Multi-Normal distribution
2. Implement a Full Bayes Classifier based on Multi-Normal distribution
3. Implement a Discrete Naive Bayes Classifier

Logistic Regression, Bayes and EM

1. Implement Logistic Regression algorithm.
2. Implement EM algorithm.
3. Implement Navie Bayes algorithm that uses EM for calculating the likelihood.
4. Visualize my results.

Clustering

1. Implement k-means as an image compression algorithm.