Following the seminal idea of Tukey (1975), data depth is a function that measures how close an arbitrary point of the space is located to an implicitly defined center of a data cloud. Having undergone theoretical and computational developments, it is now employed in numerous applications. The data-depth library is a software directed to fuse experience of the applicant with recent achievements in the area of data depth. This library provides an implementation for exact and approximate computation of most reasonable and widely applied notions of data-depth.
- Website: https://data-depth.github.io
- Source code: https://github.com/data-depth/library
- Contributing: https://data-depth.github.io/multivariate/credits.html#contributors
data-depth can be directly installed using pip:
pip install data-depth
Running GPU based depth requires CUDA availability, it can be installed using pytorch:
pip install torch --index-url https://download.pytorch.org/whl/cu118
pip install data-depth
When using conda, run before:
conda install conda-forge::pytorch
pip install data-depth
Or for GPU usage:
conda install conda-forge::pytorch-gpu
pip install data-depth
For more information about CUDA version, see https://pytorch.org/get-started/locally/