Welcome to pyTAMS’s documentation!¶
pyTAMS is a modular implementation of the trajectory adaptive multilevel splitting (TAMS), an algorithm developed to evaluate the probability of rare events associated with stochastic systems. In particular, TAMS can be used to evaluate the transition probability between two stable states of a multi-stable system. pyTAMS was developed to specifically handle (computationally) expensive stochastic models, where integrating the model can take hours to days on supercomputers and using a naive Monte-Carlo approach is impractical.
Installation:¶
To install pyTAMS, simply use pip in your favorite environment manager
to get the latest stable version:
pip install pytams
or if you plan on modifying the code or test the shipped-in examples, install from sources:
git clone git@github.com:nlesc-eTAOC/pyTAMS.git
cd pyTAMS
pip install -e .
Quick start:¶
Only if you have used the second option above you can readily test pyTAMS on a simple problem:
cd pyTAMS/examples/DoubleWell2D
python tams_dw2dim.py
otherwise, please read through this documentation and in particular follow the tutorials Section to see how to implement your own model within pyTAMS.
User guide:
Developer guide:
Documentation¶
The documentation pages are distributed with the code in the docs
folder as “reStructuredText” files. The HTML is built automatically
whenever changes are pushed to the main branch on GitHub.
A local version can also be built as follows:
cd <pyTAMS_root_folder>/docs
make html