Basic machine learning algorithms and problems.
This repository was created to realize Coursera's 2018 Machine Learning Course offered by Stanford | Online and taught by Dr. Andrew Ng (completed June 2018 with a grade of 96.07% - certificate available upon request). It has expanded slightly to explore some independent, Python-based ML.
Pull the master branch of this repository.
The full content of the MATLAB-based machine learning course from Coursera is located in the Coursera_ML folder. Here, a collection of 8 exercises are organized into folders Ex1, Ex2, etc. Each contains a collection of .m's and ocassioanlly some raw data in .txt files to accomplish some machine learning tasks. Here is a summary:
- Ex1: Linear Regression (cost function, feature normalization, gradient descent)
- Ex2: Logistic Regression and Regularization (more cost functions, sigmoid function)
- Ex3: Multiclass Classification and Neural Network Predictions (neural network units and weights, logistic regression)
- Ex4: Neural Networks and Back-Propagation (w/ and w/out regularization)
- Ex5: Regularized Linear/Polynomial Regression and Bias Variance
- Ex6: Support Vector Machines (Gaussian kernel, RBF kernel -- email processing application)
- Ex7: Principle Component Analysis and K-Means Clustering (centroids, projection -- image compression application)
- Ex8: Anomaly Detection and Collaborative Filtering (movie rating application)
A brief exploration of the "Hello, World!" of machine learning: the iris flower data set (1936).
- Matt Runyon
- Coursera (for some templating code)
- https://machinelearningmastery.com/machine-learning-in-python-step-by-step/ (for iris data set code)