Fast algorithm with dual extrapolation for the Lasso
Please visit https://mathurinm.github.io/celer/ for the latest version of the documentation.
Install the released version
Assuming you have a working Python environment, e.g. with Anaconda you can install celer with pip.
From a console or terminal install celer with pip:
pip install -U celer
To setup a fully functional environment we recommend you download this conda environment and install it with:
conda env create --file environment.yml
Install the development version
From a console or terminal clone the repository and install CELER:
git clone https://github.com/mathurinm/celer.git cd celer/ conda env create --file environment.yml source activate celer-env pip install --no-deps -e .
Demos & Examples
You find on the documentation examples on the Leukemia dataset (comparison with scikit-learn) and on the Finance/log1p dataset (more significant, but it takes times to download the data, preprocess it, and compute the path).
All dependencies are in ./environment.yml
If you use this code, please cite:
Mathurin Massias, Alexandre Gramfort and Joseph Salmon Celer: Fast solver for the Lasso with dual extrapolation To appear in ICML 2018