sci-ml/trl
Train transformer language models with reinforcement learning (SFT, DPO, GRPO)
Screenshots
View
Download
BrowseRuntime Dependencies
trl-1.7.1
>=sci-ml/accelerate-1.4.0[python_single_target_python3_12
(-)
?
,python_single_target_python3_13
(-)
?
]
>=sci-ml/datasets-4.7.0[python_single_target_python3_12
(-)
?
,python_single_target_python3_13
(-)
?
]
>=sci-ml/transformers-4.56.2[python_single_target_python3_12
(-)
?
,python_single_target_python3_13
(-)
?
]
python_single_target_python3_12?
( dev-python/jinja2[python_targets_python3_12
(-)
] >=dev-python/packaging-20.0[python_targets_python3_12
(-)
] )
python_single_target_python3_13?
( dev-python/jinja2[python_targets_python3_13
(-)
] >=dev-python/packaging-20.0[python_targets_python3_13
(-)
] )
python_single_target_python3_12?
( dev-lang/python:3.12 )
python_single_target_python3_13?
( dev-lang/python:3.13 )