Spaces:
Paused
Paused
| [metadata] | |
| name = trl | |
| version = 0.19.0.dev0 | |
| description = Train transformer language models with reinforcement learning. | |
| long_description = file: README.md | |
| long_description_content_type = text/markdown | |
| author = Leandro von Werra | |
| author_email = [email protected] | |
| url = https://github.com/huggingface/trl | |
| keywords = transformers, huggingface, language modeling, post-training, rlhf, sft, dpo, grpo | |
| license_file = LICENSE | |
| classifiers = | |
| Development Status :: 2 - Pre-Alpha | |
| Intended Audience :: Developers | |
| Intended Audience :: Science/Research | |
| Natural Language :: English | |
| Operating System :: OS Independent | |
| Programming Language :: Python :: 3 | |
| Programming Language :: Python :: 3.9 | |
| Programming Language :: Python :: 3.10 | |
| Programming Language :: Python :: 3.11 | |
| Programming Language :: Python :: 3.12 | |
| Programming Language :: Python :: 3.13 | |
| [options] | |
| packages = find: | |
| python_requires = >=3.9 | |
| include_package_data = True | |
| install_requires = | |
| accelerate>=1.4.0 | |
| datasets>=3.0.0 | |
| transformers>=4.51.0 | |
| [options.packages.find] | |
| exclude = | |
| tests* | |
| [options.package_data] | |
| trl = | |
| templates/*.md | |
| accelerate_configs/*.yaml | |
| [options.extras_require] | |
| bco = | |
| scikit-learn | |
| joblib | |
| deepspeed = | |
| deepspeed>=0.14.4 | |
| diffusers = | |
| diffusers>=0.18.0 | |
| judges = | |
| openai>=1.23.2 | |
| llm-blender>=0.0.2 | |
| liger = | |
| liger-kernel>=0.5.9 | |
| peft = | |
| peft>=0.8.0 | |
| quantization = | |
| bitsandbytes | |
| scikit = | |
| scikit-learn | |
| test = | |
| parameterized | |
| pytest-cov | |
| pytest-rerunfailures | |
| pytest-xdist | |
| pytest | |
| vllm = | |
| # vLLM package does not yet support Python 3.13. These constraints can be lifted once support is added: | |
| # see https://github.com/vllm-project/vllm/pull/13164 | |
| vllm>=0.8.3; python_version < "3.13" | |
| fastapi; python_version < "3.13" | |
| pydantic; python_version < "3.13" | |
| requests; python_version < "3.13" | |
| uvicorn; python_version < "3.13" | |
| vlm = | |
| Pillow | |
| dev = | |
| %(bco)s | |
| %(deepspeed)s | |
| %(diffusers)s | |
| %(judges)s | |
| %(liger)s | |
| %(peft)s | |
| %(quantization)s | |
| %(scikit)s | |
| %(test)s | |
| %(vlm)s | |
| [options.entry_points] | |
| console_scripts = | |
| trl = trl.cli:main | |
| [coverage:run] | |
| branch = True | |