1 files changed, 53 insertions, 0 deletions
diff --git a/sci-libs/datasets/datasets-2.12.0.ebuild b/sci-libs/datasets/datasets-2.12.0.ebuild
new file mode 100644
index 000000000000..66b609fd2b57
--- /dev/null
+++ b/sci-libs/datasets/datasets-2.12.0.ebuild
@@ -0,0 +1,53 @@
+# Copyright 2023 Gentoo Authors
+# Distributed under the terms of the GNU General Public License v2
+
+EAPI=8
+
+DISTUTILS_USE_PEP517=setuptools
+PYTHON_COMPAT=( python3_{9..11} )
+DISTUTILS_SINGLE_IMPL=1
+inherit distutils-r1
+
+DESCRIPTION="Access and share datasets for Audio, Computer Vision, and NLP tasks"
+HOMEPAGE="
+	https://pypi.org/project/datasets/
+"
+SRC_URI="https://github.com/huggingface/${PN}/archive/refs/tags/${PV}.tar.gz
+	-> ${P}.gh.tar.gz"
+IUSE="test"
+
+LICENSE="Apache-2.0"
+SLOT="0"
+KEYWORDS="~amd64"
+
+RDEPEND="
+	${PYTHON_DEPS}
+	sci-libs/pytorch[${PYTHON_SINGLE_USEDEP}]
+	$(python_gen_cond_dep '
+		dev-python/absl-py[${PYTHON_USEDEP}]
+		dev-python/aiohttp[${PYTHON_USEDEP}]
+		dev-python/fsspec[${PYTHON_USEDEP}]
+		dev-python/multiprocess[${PYTHON_USEDEP}]
+		dev-python/pandas[${PYTHON_USEDEP}]
+		dev-python/pyarrow[${PYTHON_USEDEP},parquet,snappy]
+		dev-python/tqdm[${PYTHON_USEDEP}]
+		dev-python/xxhash[${PYTHON_USEDEP}]
+		dev-python/zstandard[${PYTHON_USEDEP}]
+		sci-libs/huggingface_hub[${PYTHON_USEDEP}]
+		sci-libs/scikit-learn[${PYTHON_USEDEP}]
+	')
+"
+DEPEND="${RDEPEND}"
+BDEPEND="test? (
+	$(python_gen_cond_dep '
+		dev-python/pytest-datadir[${PYTHON_USEDEP}]
+		dev-python/decorator[${PYTHON_USEDEP}]
+		=dev-python/sqlalchemy-1*[${PYTHON_USEDEP}]
+		sci-libs/jiwer[${PYTHON_USEDEP}]
+		sci-libs/seqeval[${PYTHON_USEDEP}]
+	')
+)"
+
+PATCHES=( "${FILESDIR}"/${P}-tests.patch )
+
+distutils_enable_tests pytest