Date: Wed, 18 Mar 2026 03:00:43 +0000 From: Yuri Victorovich <yuri@FreeBSD.org> To: ports-committers@FreeBSD.org, dev-commits-ports-all@FreeBSD.org, dev-commits-ports-main@FreeBSD.org Subject: git: e6238161de04 - main - math/py-torchsde: New port: SDE solvers and stochastic adjoint sensitivity analysis in PyTorch Message-ID: <69ba155b.18890.4b38e403@gitrepo.freebsd.org>
index | next in thread | raw e-mail
The branch main has been updated by yuri: URL: https://cgit.FreeBSD.org/ports/commit/?id=e6238161de0499df239b921a2c35e7b8793ed46b commit e6238161de0499df239b921a2c35e7b8793ed46b Author: Yuri Victorovich <yuri@FreeBSD.org> AuthorDate: 2026-03-18 02:18:24 +0000 Commit: Yuri Victorovich <yuri@FreeBSD.org> CommitDate: 2026-03-18 03:00:37 +0000 math/py-torchsde: New port: SDE solvers and stochastic adjoint sensitivity analysis in PyTorch --- math/Makefile | 1 + math/py-torchsde/Makefile | 26 ++++++++++++++++++++++++++ math/py-torchsde/distinfo | 3 +++ math/py-torchsde/pkg-descr | 9 +++++++++ 4 files changed, 39 insertions(+) diff --git a/math/Makefile b/math/Makefile index 2d46f59b1b68..6f3b04131e5f 100644 --- a/math/Makefile +++ b/math/Makefile @@ -1146,6 +1146,7 @@ SUBDIR += py-theano SUBDIR += py-timple SUBDIR += py-topologic + SUBDIR += py-torchsde SUBDIR += py-triangle SUBDIR += py-umap-learn SUBDIR += py-uncertainties diff --git a/math/py-torchsde/Makefile b/math/py-torchsde/Makefile new file mode 100644 index 000000000000..19f8b360ee9d --- /dev/null +++ b/math/py-torchsde/Makefile @@ -0,0 +1,26 @@ +PORTNAME= torchsde +DISTVERSION= 0.2.6 +CATEGORIES= math python +MASTER_SITES= PYPI +PKGNAMEPREFIX= ${PYTHON_PKGNAMEPREFIX} + +MAINTAINER= yuri@FreeBSD.org +COMMENT= SDE solvers and stochastic adjoint sensitivity analysis in PyTorch +WWW= https://github.com/google-research/torchsde + +LICENSE= APACHE20 +LICENSE_FILE= ${WRKSRC}/LICENSE + +BUILD_DEPENDS= ${PY_SETUPTOOLS} \ + ${PYTHON_PKGNAMEPREFIX}wheel>0:devel/py-wheel@${PY_FLAVOR} +RUN_DEPENDS= ${PYNUMPY} \ + ${PYTHON_PKGNAMEPREFIX}pytorch>=1.6.0:misc/py-pytorch@${PY_FLAVOR} \ + ${PYTHON_PKGNAMEPREFIX}scipy>=1.5:science/py-scipy@${PY_FLAVOR} \ + ${PYTHON_PKGNAMEPREFIX}trampoline>=0.1.2:devel/py-trampoline@${PY_FLAVOR} + +USES= python +USE_PYTHON= pep517 autoplist + +NO_ARCH= yes + +.include <bsd.port.mk> diff --git a/math/py-torchsde/distinfo b/math/py-torchsde/distinfo new file mode 100644 index 000000000000..03fee0501e79 --- /dev/null +++ b/math/py-torchsde/distinfo @@ -0,0 +1,3 @@ +TIMESTAMP = 1773765834 +SHA256 (torchsde-0.2.6.tar.gz) = 81d074d3504f9d190f1694fb526395afbe4608ee43a88adb1262a639e5b4778b +SIZE (torchsde-0.2.6.tar.gz) = 48840 diff --git a/math/py-torchsde/pkg-descr b/math/py-torchsde/pkg-descr new file mode 100644 index 000000000000..641f5c9b267a --- /dev/null +++ b/math/py-torchsde/pkg-descr @@ -0,0 +1,9 @@ +TorchSDE provides stochastic differential equation (SDE) solvers with GPU +support and efficient backpropagation in PyTorch. + +Features: +- Multiple SDE solvers (Euler-Maruyama, Milstein, etc.) +- Efficient backpropagation through SDE solvers +- GPU acceleration +- Support for various noise types +- Latent SDE models for variational inferencehome | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?69ba155b.18890.4b38e403>