Skip to content

This is the official PyTorch implementation of Convex Polytope Trees

Notifications You must be signed in to change notification settings

rezaarmand/Convex_Polytope_Trees

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

Convex Polytope Trees

This repository contains an implementation of CPT The code was tested for python 3.6

How to use

You can use the following command to train CPT for the three depth paramters of 8, 9, and 10:

python run_polytop.py  --data MNIST  --depth  8 9 10  --lr 0.01  --epochs 70 --refine_epochs 50  -s 0.01  --initial_steepness 1 --polytope_sides  60  --refine

And for Connect4, depth 2

python run_polytop.py  --data connect4  --depth  2   --lr 0.01  --epochs 40 --refine_epochs 40  -s 0.01  --initial_steepness 1 --polytope_sides  40 --refine

For regression use the file run_polytop_reg.pyand for binary classification run the file run_polytop_binary_classification.py.

The datasets are also provided.

About

This is the official PyTorch implementation of Convex Polytope Trees

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages