search

malcolmgreaves / smo-fun

An efficient implementation of the Sequential Minimal Optimization algorithm for training SVMs.
Other
3 stars 1 forks source link

Regularization: Add Model Complexity Penalty #14

Open malcolmgreaves opened 8 years ago

malcolmgreaves commented 8 years ago

Standard implementations for L1, L2, elastic net regularization on the model's complexity. Also add in a functional interface for SvmDualModel => Double to "plug-n-play" w/ the cost function during learning.

malcolmgreaves commented 8 years ago

Inspiration from scikit-learn documentation: http://scikit-learn.org/stable/auto_examples/svm/plot_svm_scale_c.html