algorithmic modeling for Rhino

Hi everyone

I am trying to make myself familiar with LunchBox’s Machine Learning components. To learn the tools I have the following task in mind:

Train the algorithm so that it is able to classify weather a Mesh is a Sphere or a Box.

I got something going and it might be working. For the training and testing data I am using lists containing the Meshes vertexes positions in space. However all my training and testing lists have to be exactly the same length otherwise I am getting an error from the Neural Network. But I would like for the meshes to have different vertex counts for this to work.

I think there might be a mistake in my data management, but I am not sure. Does anybody have any suggestions or clever ideas?



Views: 48






  • Add Photos
  • View All

© 2019   Created by Scott Davidson.   Powered by

Badges  |  Report an Issue  |  Terms of Service