Divvy is a 64-bit Mac OS X 10.6, 10.7, or 10.8 (Snow Leopard, Lion, or Mountain Lion) application for performing unsupervised machine learning and visualization. We focus on the clustering (separating data into groups) and dimensionality reduction (finding low dimensional structure in high dimensional data) subfields of machine learning. For visualization we provide support for both the whole dataset (e.g. a scatter plot) and points (e.g. transforming a particular point into an image).
Free and open. The NSF funded this work so that you don't have to. Go get our MIT-licensed code at GitHub. https://github.com/jmlewis/divvy
Endlessly extensible. Every clusterer, reducer, point visualizer and dataset visualizer in Divvy is a plugin. We've provided a few big ones (K-means, PCA, scatter plot, &c.) and we're hoping the community will use our plugin protocol to build many more. Each plugin defines its own UI, so your algorithm can look and behave the way that you want it to without top-down constraints.
Part of your workflow. Export your clusterings and reductions to .csv and your visualizations to .png. Use your R or Matlab data with our R and Matlab to Divvy export tools.
- Changes to previous version:
Initial Announcement on mloss.org.
No one has posted any comments yet. Perhaps you'd like to be the first?
Leave a comment
You must be logged in to post comments.