Merge pull request #22 from HiDiHlabs/gpu_eigendecomp

Add GPU support

Author: niklasmueboe

.gitignore

@@ -0,0 +1,11 @@
+.venv
+.python-version
+*.lock
+multispaeti.egg-info
+__pycache__
+
+.vscode
+
+# sphinx
+generated
+build

.pre-commit-config.yaml

@@ -24,6 +24,8 @@ repos:
     rev: v1.15.0
     hooks:
       - id: mypy
+        additional_dependencies:
+          - "numpy"
   - repo: https://github.com/codespell-project/codespell
     rev: v2.4.1
     hooks:

.readthedocs.yaml

@@ -2,7 +2,7 @@ version: 2
 build:
   os: ubuntu-22.04
   tools:
-    python: "3.10"
+    python: "3.11"
 sphinx:
   configuration: docs/source/conf.py
 python:

docs/source/conf.py

@@ -50,6 +50,7 @@
 exclude_patterns: list[str] = []
 
 intersphinx_mapping = dict(
+    cupy=("https://docs.cupy.dev/en/stable/", None),
     matplotlib=("https://matplotlib.org/stable/", None),
     numpy=("https://numpy.org/doc/stable/", None),
     python=("https://docs.python.org/3", None),

docs/source/installation.rst

@@ -11,6 +11,23 @@ To install ``multispaeti`` from `PyPI <https://pypi.org/>`_ using ``pip`` just r
 
     pip install multispaeti
 
+GPU support
+___________
+
+In most cases support for GPU computations (using `CuPy <https://cupy.dev/>`_) can be
+installed via
+
+.. code-block:: bash
+
+    # for CUDA12
+    pip install multispaeti[cuda12]
+
+    # or for CUDA11
+    pip install multispaeti[cuda11]
+
+However, in cases where this doesn't work we recommended referring to the
+`CuPy documentation <https://docs.cupy.dev/en/stable/install.html>`_.
+
 
 conda-forge and ``conda``
 -------------------------
@@ -26,6 +43,12 @@ conda-forge and ``conda``
     Of course, it is also possible to use ``mamba`` instead of ``conda``
     to speed up the installation.
 
+GPU support
+___________
+
+For support of GPU computations please refer to the
+`CuPy documentation <https://docs.cupy.dev/en/stable/install.html>`_.
+
 
 From GitHub
 -----------