pyensmallen: python bindings for the ensmallen library for numerical optimization

Very minimal python bindings for ensmallen library. Currently supports

• L-BFGS, with intended use for optimisation of smooth objectives for m-estimation
• ADAM (and variants with different step-size routines) - makes use of ensmallen's templatization.
• Frank-Wolfe, with intended use for constrained optimization of smooth losses
  • constraints are either lp-ball (lasso, ridge, elastic-net) or simplex

See ensmallen docs for details.

Installation:

from pypi

pip install pyensmallen

from source

1. Install armadillo and ensmallen for your system (build from source, or via conda-forge; I went with the latter)
2. git clone this repository
3. pip install -e .
4. Profit? Or at least minimize loss?

from wheel

• download the appropriate .whl for your system from the more recent release listed in Releases and run pip install ./pyensmallen... OR
• copy the download url and run pip install https://github.com/apoorvalal/pyensmallen/releases/download/<version>/pyensmallen-<version>-<pyversion>-linux_x86_64.whl