4 ms·
Good to see others looking into tree model viz. I've done work with larger scale tree visualizations and found you quickly run out of space. I wound up using
by jdonaldson 8y ago
Good to see others looking into tree model viz. I've done work with larger scale tree visualizations and found you quickly run out of space. I wound up using interactivity to reveal branch level info, dynamically pruned the tree based on train support, and I used a more sophisticated layout technique to pack more info in. https://www.google.com/amp/s/blog.bigml.com/2012/01/23/beautiful-decisions-inside-bigmls-decision-trees/amp/ https://www.google.com/amp/s/blog.bigml.com/2012/01/23/beaut...
FWIW visualizing trees like that helps spot problems really quickly. Overfitting behavior typically involves overusing a certain field, or growing long and relatively narrow branches.
- parrt 8y agoThanks for that link. Super useful. Looks like BigML uses same layout I did for ANTLR parse trees. Really packs stuff in; e.g., https://cdn-images-1.medium.com/max/1760/1*k0mO4kJyQvPCyyev0nn_EQ.png https://cdn-images-1.medium.com/max/1760/1*k0mO4kJyQvPCyyev0...
- jdonaldson 8y agoYeah the general algorithm is "Reingold-Tilford" with some tweaks from Buchehim. https://en.m.wiktionary.org/wiki/Reingold-Tilford_algorithm https://en.m.wiktionary.org/wiki/Reingold-Tilford_algorithm Really good algorithm to have in an arbitrary visualization toolkit.