Metropolis-adjusted Langevin Algorithm (MALA] is a Markov Chain Monte Carlo (MCMC) method to sample from intractable probability distributions. This algorithm proposes states using overdamped Langevin dynamics, and accepts/rejects them using the Metropolis-Hastings algorithm. The former directs the random walk towards high probability regions, while the latter is used to avoid localization and promote mixing.
I will be implementing this algorithm using ASE's Langevin class.
To install from the source directory, run
pip install .
To install optional dependencies, such as examples, run the command
pip install .[examples]
To build this package locally, run the following command in the main project directory
pip -m pip install build
python -m build
Then, to install the package,
pip install dist/*.whl