tyler-tomita / RandomerForest

Discriminant Projection Forest results, datasets, etc.
44 stars 21 forks source link

Update Randomer Forest so that only the split dimension for each node is stored #36

Closed tyler-tomita closed 9 years ago

tyler-tomita commented 9 years ago

Done. Makes a world of difference. To give you an idea, for Trunk n=100, d=1000, an instantiation of dense randomer forest is only about 4x larger than that of random forest. Previously, it was a few orders of magnitude larger.

jovo commented 9 years ago

nice! I presume performance is the same?

On Thursday, April 2, 2015, ttomita notifications@github.com wrote:

Done. Makes a world of difference. To give you an idea, for Trunk n=100, d=1000, an instantiation of dense randomer forest is only about 4x larger than that of random forest. Previously, it was a few orders of magnitude larger.

— Reply to this email directly or view it on GitHub https://github.com/ttomita/DPForest/issues/36.

the glass is all full: half water, half air. openconnecto.me, jovo.me, my calendar https://www.google.com/calendar/embed?src=joshuav%40gmail.com&ctz=America/New_York