online-learning

I've been doing machine learning research in the area of online classification. As great as the Python tools are for batch learning, there seems to be a real lack of online learning libraries, so I decided I would contribute my implementations of several online learning algorithms I've been using in my research as a start at building a Python library for this type of machine learning.

I hope to build this out into into a full library for online learning, so if you're interested in helping out or have any ideas for the project, don't hesitate to get in touch! The starting work here is based off of the Matlab and C++ implementations found in LIBOL. The following alogithms are currently included:

1. Passive Aggressive Algorithms
   • Online Passive-Aggressive Algorithms
2. Random Budgeted Perceptron
   • Tracking the best hyperplane with a simple budget Perceptron
3. Fourier Online Gradient Descent
   • Large Scale Online Kernel Learning
4. Dual Space Gradient Descent
   • Dual Space Gradient Descent for Online Learning