1
0
Fork
You've already forked ctmc
0
Continuous Time Monte Carlo on arbitrary graphs... eventually
  • Rust 100%
2026年04月19日 22:37:32 +05:30
papers add papers 2026年04月19日 22:36:52 +05:30
src fix imports 2026年04月19日 22:37:32 +05:30
tests impl ising model, square lattice, metropolis 2026年04月17日 18:37:50 +05:30
.gitignore housekeeping 2026年04月19日 21:50:16 +05:30
9802104v1.pdf add README.md and the main paper 2026年04月19日 22:14:10 +05:30
Cargo.lock housekeeping 2026年04月19日 21:50:16 +05:30
Cargo.toml housekeeping 2026年04月19日 21:50:16 +05:30
README.md add README.md and the main paper 2026年04月19日 22:14:10 +05:30
rust-toolchain.toml impl ising model, square lattice, metropolis 2026年04月17日 18:37:50 +05:30
TODO.md rename ising_metropolis.rs to metropolis.rs 2026年04月19日 21:51:20 +05:30

Continuous Time Monte Carlo... IN (削除) SPACE (削除ここまで) rust. i mean rust. for some reason.

The algorithm this package wants to implement is outlined in the paper by Rieger Kawashima which allows one to simulate an Ising model near T = 0. The reason classical methods for doing so fail is due to the fact that the Suzuki Trotter decomposition makes Z = 0 at T = 0. I'll write more about it later, but you can go read the paper, it's on the archive too.

Currently there's a library and a binary. The library has src/hlattice.rs for constructing a hierarchical lattice. Formalizing this interface will be one of the last things I'll do before tagging a release. The algorithms to act on these lattice (all Ising model specific) are in src/algos (I should rename that) and there's some working simulation example accessible via the simulations binary artifact. You can check TODO.md to see where my head's at.