add LLM training scripts

tasks/generative-ai/text-to-text/training/.gitignore

@@ -0,0 +1,165 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/
+
+tmp
+*.tar.gz
+*.bin
+*.pt

tasks/generative-ai/text-to-text/training/README.md

@@ -0,0 +1,37 @@
+# SageMaker での日本語 LLM の分散学習
+
+こちらのサンプルコードは [SageMaker Model Parallel Library](https://docs.aws.amazon.com/sagemaker/latest/dg/model-parallel.html) を使用して SageMaker 上で日本語大規模言語モデルの Pretraining、Fine-Tuning、Instruction-Tuning が行うサンプルコードです。
+
+## Notebooks
+
+- smp-train-jp-gpt-neox-sharded-data-parallel.ipynb
+  - 日本語 GPT Neox の Pretraining
+- smp-finetune-jp-gpt-neox-sharded-data-parallel.ipynb
+  - 日本語 GPT Neox (Rinna 3.6B) の Fine Tuning
+- smp-instruct-jp-gpt-neox-sharded-data-parallel.ipynb
+  - 日本語 GPT Neox (Rinna 3.6B) の Instruction Tuning
+- smp-train-gpt-neox-sharded-data-parallel.ipynb
+  - GPT NeoX の Pretraining（元のノートブックを少し改変したもの）
+- convert-hf.ipynb
+  - 分散学習スクリプトで保存した Model を HuggingFace 形式に変換して保存するコード
+- data-preprocess.ipynb
+  - 事前学習用にデータセットを前処理して FSx for Lustre に保存するサンプルコード
+
+## 元サンプルコードからの変更点
+
+このサンプルノートブックは [SageMaker の分散学習での GPT NeoX の Pretraining サンプル](https://github.com/aws/amazon-sagemaker-examples/tree/main/training/distributed_training/pytorch/model_parallel/gpt-neox) を元に日本語対応および Fine-tuning、Instruction Tuning 対応などを追加しています。
+
+主な変更点は以下になります。
+
+- 日本語対応
+- Finetuning、Instruction Tuning ノートブックの追加
+- Padding Collator の実装
+  - Pretraining であればテキストをチャンク化するためバッチ内でのトークン長が一致しているため問題にならないが Instruction Tuning の際には Instruction ごとにトークン長が異なるため Padding が必要になる。[Transformers の Collator](https://github.com/huggingface/transformers/blob/main/src/transformers/data/data_collator.py#L402) の実装を元にバッチに Left Padding を施す変更を加えた。元々トークン長が一致している Pretraining には影響しない。
+- その他、試行錯誤しやすいように細かな変更をいくつか追加しています
+
+## おすすめのリソース
+
+- [大規模言語モデルを Amazon SageMaker 上で学習する際のベストプラクティス](https://aws.amazon.com/jp/blogs/news/training-large-language-models-on-amazon-sagemaker-best-practices/)
+- [SageMaker 分散モデル並列処理のベストプラクティス](https://docs.aws.amazon.com/ja_jp/sagemaker/latest/dg/model-parallel-best-practices.html)
+- [GPT-J の分散学習についての AWS ブログ](https://aws.amazon.com/blogs/machine-learning/fine-tune-gpt-j-using-an-amazon-sagemaker-hugging-face-estimator-and-the-model-parallel-library/)
+- [Amazon SageMaker 分散学習ドキュメント](https://docs.aws.amazon.com/ja_jp/sagemaker/latest/dg/distributed-training.html)