Skip to content

echanj/EVPT_2D_toy_models

Repository files navigation

Extended Variable Parallel Tempering Toy models

Author: Eric J. Chan

Date(repo created): 3/5/2023 Date(began coding): ~ 1/3/2023

Description:

Python notebokk files which demonstrate stepwise development of the Extended Variable Parallel Tempering Method which I tested and used to perform CSP sampling in the 7th blind test. Python note book files that demonstrate stepwise development of the Extended Variable Parallel Tempering (EVPT) method that I tested and used to perform a lot of CSP sampling in the 7th blind test I am currently still developing this method as a means of biassing a CSP search for elucidating crystal structure models from PXRD without the need for unit-cell indexing. The work remains unfunded and has been placed as a low priority.

2D_double_well.ipynb - check the double well potential is working basin_hop_legacy_test.ipynb  - original basin hop algoritham construction test ground, defined later in .py modules basin_hop_local_minima.ipynb - running BH on surface with many local minima      basin_hop_rough_surface.ipynb - more test with BH. this is an important notebook that also demonstrates the quasi-random and pseudo-random methods with the same PES PT_rough_surface.ipynb  - just vanilla PT on the 2D- surface, then we make the jump to PT with BH   EVPT_rough_surface.ipynb      - this is the EVPT algoritham as a toy model

associated module files:

optimize_functions.py energy_functions.py   parallel_temper_BH.py parallel_temper.py      parallel_temper1D.py parallel_temper_EVBH.py run.py

About

These are codes of toy physics models that contain building blocks for understanding the concept behind EVCCPMC

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published