py-resize-right: new package (#47056)

Co-authored-by: Alex C Leute <acl2809@rit.edu>

1 files changed, 32 insertions, 0 deletions

diff --git a/var/spack/repos/builtin/packages/py-resize-right/package.py b/var/spack/repos/builtin/packages/py-resize-right/package.py
new file mode 100644
index 0000000000..4e737f2d83
--- /dev/null
+++ b/var/spack/repos/builtin/packages/py-resize-right/package.py
@@ -0,0 +1,32 @@
+# Copyright 2013-2024 Lawrence Livermore National Security, LLC and other
+# Spack Project Developers. See the top-level COPYRIGHT file for details.
+#
+# SPDX-License-Identifier: (Apache-2.0 OR MIT)
+
+
+from spack.package import *
+
+
+class PyResizeRight(PythonPackage):
+    """This is a resizing packge for images or tensors, that supports both
+    Numpy and PyTorch (fully differentiable) seamlessly. The main motivation
+    for creating this is to address some crucial incorrectness issues (see item
+    3 in the list below) that exist in all other resizing packages I am
+    aware of. As far as I know, it is the only one that performs correctly
+    in all cases.  ResizeRight is specially made for machine learning,
+    image enhancement and restoration challenges."""
+
+    homepage = "https://github.com/assafshocher/ResizeRight"
+    pypi = "resize-right/resize-right-0.0.2.tar.gz"
+
+    license("MIT", checked_by="alex391")
+
+    version("0.0.2", sha256="7dc35b72ce4012b77f7cc9049835163793ab98a58ab8893610fb119fe59af520")
+
+    depends_on("py-setuptools", type="build")
+    # needs py-numpy, py-torch, or both. py-numpy is lighter to install, so
+    # always use py-numpy
+    depends_on("py-numpy", type=("build", "run"))
+    # and optionally use py-torch
+    variant("torch", default=True, description="Enable py-torch")
+    depends_on("py-torch", type=("build", "run"), when="+torch")