Machine Learning

This is a folder of common machine learning algorithms derived and implemented in python.

Algorithms

Supervised

This folder contains supervised learning algorithms

Classifiers

Identifying which categories a feature set belongs to

Boosting

Combining several weak classifiers into a single strong one

Adaboost

Weighted sum of individual classifiers

Linear Classifiers

Combining feature vectors with weight vectors to classify data points

SVM

Create a set of features (support vectors) to help define the decision boundary

Perceptron

Iteratively update a decision boundary based on the classification of each point. *Does not converge if not linearly seperable.

Neural Networks

Network of perceptrons feeding into one another

ANN

Straightforward Neural Network

Regression

Estimating relationship between inputs and their outputs

Unsupervised

This folder contains unsupervised learning algorithms

Clustering

Grouping Similar Objects

K-Means

Using the Mean of all points in a cluster to update the center

K-Medians

Using the Median of all points in a cluster to update the center

Supplements

Validation

Nested Cross Validation

Method for determining the strength of a model