Upgrading Sagemaker docs and tutorials to sagemaker-sdk v3#2556
Conversation
fix: removed unecessary dependencies from the `run_mnr.py` script because now we're using an image that already carries most of them change: asset folder renamed to scripts because it only contains a script and a requirements file (furthermore the notebook was pointing to the scrips/ folder and not to assets/) Signed-off-by: DWarez <dario.salvati@huggingface.co>
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 90d10a8. Configure here.
change: upgraded packages involved change: added requirements and launch script directly in a scripts/ folder, removed the instructions to write them at runtime Signed-off-by: DWarez <dario.salvati@huggingface.co>
Signed-off-by: DWarez <dario.salvati@huggingface.co>
deepseek-ocr-sagemaker Signed-off-by: DWarez <dario.salvati@huggingface.co>
Signed-off-by: DWarez <dario.salvati@huggingface.co>
Signed-off-by: DWarez <dario.salvati@huggingface.co>
Signed-off-by: DWarez <dario.salvati@huggingface.co>
Signed-off-by: DWarez <dario.salvati@huggingface.co>
ehcalabres
left a comment
There was a problem hiding this comment.
Hey @dwarez! Thanks for all the updates! As per the changes you made, everything looks good to me. I just have a couple of comments regarding unchanged sections:
- Do we want to include the note about sagemaker v3 on all notebooks? If so, we should add it here and here too
- I think that the neuron agent example is still in the old SDK version, maybe you missed that one?
Signed-off-by: DWarez <dario.salvati@huggingface.co>
hey @ehcalabres thanks for the review 😄 You're right the banner was missing from the fine-tune-trl notebook, I added it. |

Note
Low Risk
Changes are docs and notebooks only; they do not alter runtime libraries, though incorrect SDK v3 examples could mislead users configuring AWS training or endpoints.
Overview
This PR refreshes Hugging Face on AWS SageMaker documentation and example notebooks for SageMaker Python SDK v3, replacing v2 patterns (
HuggingFace/HuggingFaceModel) withModelTrainerfor jobs andModelBuilder(orsagemaker.core.resourceswhere Neuron/TGI builders do not apply) for inference.Tutorials under
docs/sagemaker/source/now call outpip install "sagemaker>=3.0.0",image_uris.retrieve,InputDatachannels, and updated deploy/train flows (including JumpStart viaModelBuilder(model=...)). Example notebooks cover embeddings (TEI), Lighteval onModelTrainer, TRL CLI SFT, DeepSeek OCR batch jobs, Inferentia2 Llama via coreModel/EndpointAPIs, and fine-tune-then-deploy embedding workflows..gitignoreadds.ipynb_checkpoints/so local Jupyter checkpoint dirs from the updated notebooks are not committed.Reviewed by Cursor Bugbot for commit 847d078. Bugbot is set up for automated code reviews on this repo. Configure here.