3 files changed, 54 insertions, 0 deletions

diff --git a/math/py-gym/Makefile b/math/py-gym/Makefile
new file mode 100644
index 000000000000..e293820443d3
--- /dev/null
+++ b/math/py-gym/Makefile
@@ -0,0 +1,27 @@
+# Created by: Po-Chuan Hsieh <sunpoet@FreeBSD.org>
+# $FreeBSD$
+
+PORTNAME=	gym
+PORTVERSION=	0.12.5
+CATEGORIES=	math python
+MASTER_SITES=	CHEESESHOP
+PKGNAMEPREFIX=	${PYTHON_PKGNAMEPREFIX}
+
+MAINTAINER=	sunpoet@FreeBSD.org
+COMMENT=	OpenAI toolkit for developing and comparing your reinforcement learning agents
+
+LICENSE=	MIT
+
+RUN_DEPENDS=	${PYTHON_PKGNAMEPREFIX}numpy>=1.10.4:math/py-numpy@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}pyglet>=0:graphics/py-pyglet@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}scipy>=0:science/py-scipy@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}six>=0:devel/py-six@${PY_FLAVOR}
+TEST_DEPENDS=	${PYTHON_PKGNAMEPREFIX}mock>=0:devel/py-mock@${PY_FLAVOR} \
+		${PYTHON_PKGNAMEPREFIX}pytest>=0:devel/py-pytest@${PY_FLAVOR}
+
+USES=		python
+USE_PYTHON=	autoplist concurrent distutils
+
+NO_ARCH=	yes
+
+.include <bsd.port.mk>
diff --git a/math/py-gym/distinfo b/math/py-gym/distinfo
new file mode 100644
index 000000000000..f57e8cf90bb2
--- /dev/null
+++ b/math/py-gym/distinfo
@@ -0,0 +1,3 @@
+TIMESTAMP = 1561148961
+SHA256 (gym-0.12.5.tar.gz) = 027422f59b662748eae3420b804e35bbf953f62d40cd96d2de9f842c08de822e
+SIZE (gym-0.12.5.tar.gz) = 1544308
diff --git a/math/py-gym/pkg-descr b/math/py-gym/pkg-descr
new file mode 100644
index 000000000000..291faba27a40
--- /dev/null
+++ b/math/py-gym/pkg-descr
@@ -0,0 +1,24 @@
+OpenAI Gym is a toolkit for developing and comparing reinforcement learning
+algorithms. This is the gym open-source library, which gives you access to a
+standardized set of environments.
+
+gym makes no assumptions about the structure of your agent, and is compatible
+with any numerical computation library, such as TensorFlow or Theano. You can
+use it from Python code, and soon from other languages.
+
+There are two basic concepts in reinforcement learning: the environment (namely,
+the outside world) and the agent (namely, the algorithm you are writing). The
+agent sends actions to the environment, and the environment replies with
+observations and rewards (that is, a score).
+
+The core gym interface is Env, which is the unified environment interface. There
+is no interface for agents; that part is left to you. The following are the Env
+methods you should know:
+- reset(self): Reset the environment's state. Returns observation.
+- step(self, action): Step the environment by one timestep. Returns observation,
+  reward, done, info.
+- render(self, mode='human'): Render one frame of the environment. The default
+  mode will do something human friendly, such as pop up a window.
+
+WWW: https://gym.openai.com/
+WWW: https://github.com/openai/gym