Computational Verification of the Cone Conjecture v1.0

Written and tested in Ubuntu 16.04.4 LTS (xenial), using the development branch of SAGE also.

This is a computational experiment, implementing the Top Down and Bottom Up algorithms found in "The Poset of Rational Cones" (https://msp.org/pjm/2018/292-1/p05.xhtml)

Getting Started

Dependencies

SAGE - Make sure to compile SAGE from source so that custom packages can be used! Download the developer branch from git.

https://github.com/sagemath/sage

Similarly, make Normaliz and PyNormaliz. Need to document the installation process for getting the development branch pyNormaliz2

Check ticket #25090 https://trac.sagemath.org/ticket/25090 Once closed.

1. Update sage by navigating to the sage folder, then do the following four commands:

1)	git checkout develop
2)	make
3)	sage -i normaliz
4)	sage -i pynormaliz

Check when a ticket is merged in which version: https://groups.google.com/forum/#!forum/sage-release

Instructions

Install dependencies, save the .py files listed below to the same directory. Then run from terminal in that directory:

chmod +x cone_conjecture_tester.py
./cone_conjecture_tester.py

And have fun~

The data will be outputted into the folder "DATA/{}d/experiment_name" where {} is the dimension of the experiment, and experiment_name is what you choose in the UI.

Main .json file will be named the same as the experiment name. Data Summary.text - read the title. Graphs : "SIZE" - the number of hilbert basis elements. "LENGTH" - the euclidean norm of the longest element of the hilbert basis element.

Modules

cone_conjecture_tester.py

Main UI for interfacing with ConeChain objects, and allows user to create, load, generate and save data.

cone_chain.py

Data structure to store the possible poset chains and the algorithms top_down and bottom_up.

cone_chain_element.py

Data structure to hold a polyhedral cone (sage.all.Polyhedra) and holds the hilbert basis of said cone, so that the calculation needs to only be done once.

cone_tools.py

Holds functions that are used in the top_down and bottom_up algorithms

experiment_io_tools.py

Holds I/O functions that simplify coding work for cone_conjecture_tester, such as custom menus and user input functions.