Machine Learning, Stanford
This repository contains the most recent versions of all projects and peer assessments for the Machine Learning, Andrew Ng Coursera course.
1. Introduction to Data Science in Python
2. Linear Regression
Implement the multivariate linear regression on house pricing data and plot the contour of cost function.
3. Logistic Regression
Implement the logistic regression for quality assurance of microchips from a fabrication plant. The model is further improved using regularization techniques.
4. Neural Networks: Representation
Implement and compare multi-class logistic regression and neural network over hand-written MNIST dataset. Conclude the advantage of neural network by forming non-linear hypothesis.
5. Neural Networks: Learning
Implement the same neural network over MNIST dataset while digging the concept of backpropagation and effect of regularization.
6. Applying Machine Learning
Build model with different parameteric values and study the resulting bias-variance tradeoff.
7. Support Vector Machines
Build model to get deep understanding of support vector machines b implementing it on spam email classification dataset.
8. Unsupervised Learning
Implement K-means algorithm for image compression by substituting the whole set of colors used in the image by set of color of K centroids. Also, implement the PCA on face dataset for the dimensionality reduction.
9. Anomaly Detection
Implement the anomaly detection algorithm on server computers by fitting gaussian distribution over dataset. Also, implement the collaborative filtering learning algorithm on movie ratings dataset.