some ideas of things to implement: 

main functionality
- supervised learning
  - classification
    - linear classifiers (e.g. naive bayes, perceptron, etc.)
    - quadratic classifiers
    - support vector machines
    - kernel estimation (e.g. k-nearest neighbors)
    - decision trees & random forests
    - neural networks
    - genetic programming
  - regression
    - linear/logistic regression
  - 
- unsupervised learning
  - Dimensionality reduction (e.g. Principle component analysis)
  - clustering (e.g. k-means)
  - natural language processing




supporting modules:
- statistics