This repository contains a few Machine Learning hacks written in Torch7 (or Julia). Most of the scripts implement basic machine learning algorithms or follow proofs from Christopher Bishop's "Pattern Recognition and Machine Learning".
The biggest part of the code is written in lua modules, but some proofs are written in iTorch notebooks.
The code is written in weekends so there's a slow pace in adding new algorithms.
- a very flexible module that generates synthetic data sets for you
ds = require('synthetic_data_sets.lua')
ds.random_data()
args = {['D'] = 2, ['K'] = 2, ['uniform'] = true}
ds.random_data(args)
- miscellaneous
- a short demonstration of the value iteration algorithm