1: \begin{abstract}

2:   We introduce \Geomstats, an open-source Python toolbox for computations and

3:   statistics on nonlinear manifolds, such as hyperbolic spaces, spaces of

4:   symmetric positive definite matrices, Lie groups of transformations, and many

5:   more.

6:   We provide object-oriented and extensively unit-tested implementations.

7:   Among others, manifolds come equipped with families of Riemannian metrics,

8:   with associated exponential and logarithmic maps, geodesics and parallel

9:   transport.

10:   Statistics and learning algorithms provide methods for estimation, clustering

11:   and dimension reduction on manifolds.

12:   All associated operations are vectorized for batch computation and provide

13:   support for different execution backends, namely NumPy, PyTorch and

14:   TensorFlow, enabling GPU acceleration.

15:   This paper presents the package, compares it with related libraries and

16:   provides relevant code examples.

17:   We show that \Geomstats provides reliable building blocks

18:   to foster research in differential geometry and statistics, and to

19:   democratize the use of Riemannian geometry in machine learning applications.

20:   The source code is freely available under the MIT license at

21:   \Href{http://geomstats.ai}{geomstats.ai}.

22: \end{abstract}

23: