Add setup.py wheel building option

[pseudotensor]

(h2ollm) jon@pseudotensor:~/h2ogpt$ GPU=1 python setup.py sdist bdist_wheel
running sdist
running egg_info
writing h2ogpt.egg-info/PKG-INFO
writing dependency_links to h2ogpt.egg-info/dependency_links.txt
writing requirements to h2ogpt.egg-info/requires.txt
writing top-level names to h2ogpt.egg-info/top_level.txt
reading manifest file 'h2ogpt.egg-info/SOURCES.txt'
adding license file 'LICENSE'
writing manifest file 'h2ogpt.egg-info/SOURCES.txt'
running check
creating h2ogpt-0.1.0
creating h2ogpt-0.1.0/h2ogpt.egg-info
creating h2ogpt-0.1.0/tests
copying files to h2ogpt-0.1.0...
copying LICENSE -> h2ogpt-0.1.0
copying README.md -> h2ogpt-0.1.0
copying setup.py -> h2ogpt-0.1.0
copying ./cli.py -> h2ogpt-0.1.0/.
copying ./client_test.py -> h2ogpt-0.1.0/.
copying ./create_data.py -> h2ogpt-0.1.0/.
copying ./eval.py -> h2ogpt-0.1.0/.
copying ./export_hf_checkpoint.py -> h2ogpt-0.1.0/.
copying ./finetune.py -> h2ogpt-0.1.0/.
copying ./generate.py -> h2ogpt-0.1.0/.
copying ./gpt4all_llm.py -> h2ogpt-0.1.0/.
copying ./gpt_langchain.py -> h2ogpt-0.1.0/.
copying ./gradio_runner.py -> h2ogpt-0.1.0/.
copying ./gradio_themes.py -> h2ogpt-0.1.0/.
copying ./h2oai_pipeline.py -> h2ogpt-0.1.0/.
copying ./image_captions.py -> h2ogpt-0.1.0/.
copying ./llama_flash_attn_monkey_patch.py -> h2ogpt-0.1.0/.
copying ./loaders.py -> h2ogpt-0.1.0/.
copying ./make_db.py -> h2ogpt-0.1.0/.
copying ./pipeline_example.py -> h2ogpt-0.1.0/.
copying ./prompter.py -> h2ogpt-0.1.0/.
copying ./read_wiki_full.py -> h2ogpt-0.1.0/.
copying ./stopping.py -> h2ogpt-0.1.0/.
copying ./utils.py -> h2ogpt-0.1.0/.
copying h2ogpt.egg-info/PKG-INFO -> h2ogpt-0.1.0/h2ogpt.egg-info
copying h2ogpt.egg-info/SOURCES.txt -> h2ogpt-0.1.0/h2ogpt.egg-info
copying h2ogpt.egg-info/dependency_links.txt -> h2ogpt-0.1.0/h2ogpt.egg-info
copying h2ogpt.egg-info/requires.txt -> h2ogpt-0.1.0/h2ogpt.egg-info
copying h2ogpt.egg-info/top_level.txt -> h2ogpt-0.1.0/h2ogpt.egg-info
copying tests/test_cli.py -> h2ogpt-0.1.0/tests
copying tests/test_client_calls.py -> h2ogpt-0.1.0/tests
copying tests/test_eval.py -> h2ogpt-0.1.0/tests
copying tests/test_eval_models.py -> h2ogpt-0.1.0/tests
copying tests/test_imports.py -> h2ogpt-0.1.0/tests
copying tests/test_langchain_simple.py -> h2ogpt-0.1.0/tests
copying tests/test_langchain_units.py -> h2ogpt-0.1.0/tests
copying tests/test_manual_test.py -> h2ogpt-0.1.0/tests
copying tests/test_prompter.py -> h2ogpt-0.1.0/tests
copying tests/test_requirements.py -> h2ogpt-0.1.0/tests
Writing h2ogpt-0.1.0/setup.cfg
Creating tar archive
removing 'h2ogpt-0.1.0' (and everything under it)
running bdist_wheel
running build
running build_py
/home/jon/miniconda3/envs/h2ollm/lib/python3.10/site-packages/setuptools/_distutils/cmd.py:66: SetuptoolsDeprecationWarning: setup.py install is deprecated.
!!

        ********************************************************************************
        Please avoid running ``setup.py`` directly.
        Instead, use pypa/build, pypa/installer, pypa/build or
        other standards-based tools.

        See https://blog.ganssle.io/articles/2021/10/setup-py-deprecated.html for details.
        ********************************************************************************

!!
  self.initialize_options()
installing to build/bdist.linux-x86_64/wheel
running install
running install_lib
creating build/bdist.linux-x86_64/wheel
creating build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/h2oai_pipeline.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/create_data.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/loaders.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/eval.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/llama_flash_attn_monkey_patch.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/gradio_themes.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/finetune.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/read_wiki_full.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/stopping.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/image_captions.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/utils.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/gradio_runner.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/generate.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/cli.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/gpt_langchain.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/prompter.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/gpt4all_llm.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/pipeline_example.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/client_test.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/export_hf_checkpoint.py -> build/bdist.linux-x86_64/wheel/h2ogpt
copying build/lib/h2ogpt/make_db.py -> build/bdist.linux-x86_64/wheel/h2ogpt
running install_egg_info
Copying h2ogpt.egg-info to build/bdist.linux-x86_64/wheel/h2ogpt-0.1.0-py3.10.egg-info
running install_scripts
creating build/bdist.linux-x86_64/wheel/h2ogpt-0.1.0.dist-info/WHEEL
creating 'dist/h2ogpt-0.1.0-py3-none-any.whl' and adding 'build/bdist.linux-x86_64/wheel' to it
adding 'h2ogpt/cli.py'
adding 'h2ogpt/client_test.py'
adding 'h2ogpt/create_data.py'
adding 'h2ogpt/eval.py'
adding 'h2ogpt/export_hf_checkpoint.py'
adding 'h2ogpt/finetune.py'
adding 'h2ogpt/generate.py'
adding 'h2ogpt/gpt4all_llm.py'
adding 'h2ogpt/gpt_langchain.py'
adding 'h2ogpt/gradio_runner.py'
adding 'h2ogpt/gradio_themes.py'
adding 'h2ogpt/h2oai_pipeline.py'
adding 'h2ogpt/image_captions.py'
adding 'h2ogpt/llama_flash_attn_monkey_patch.py'
adding 'h2ogpt/loaders.py'
adding 'h2ogpt/make_db.py'
adding 'h2ogpt/pipeline_example.py'
adding 'h2ogpt/prompter.py'
adding 'h2ogpt/read_wiki_full.py'
adding 'h2ogpt/stopping.py'
adding 'h2ogpt/utils.py'
adding 'h2ogpt-0.1.0.dist-info/LICENSE'
adding 'h2ogpt-0.1.0.dist-info/METADATA'
adding 'h2ogpt-0.1.0.dist-info/WHEEL'
adding 'h2ogpt-0.1.0.dist-info/top_level.txt'
adding 'h2ogpt-0.1.0.dist-info/RECORD'
removing build/bdist.linux-x86_64/wheel

and in fresh env:

(h2ogpt4) jon@pseudotensor:~/h2ogpt$ pip install dist/h2ogpt-0.1.0-py3-none-any.whl --extra-index-url https://download.pytorch.org/whl/cu117

then run

python generate.py --base_model=h2oai/h2ogpt-oig-oasst1-512-6_9b --load_8bit=True

and works.