Update container URIs #1056
Update container URIs #1056
Conversation
dacorvo
requested review from
JingyaHuang,
Copilot,
michaelbenayoun,
pagezyhf and
tengomucho
There was a problem hiding this comment.
Pull request overview
This PR updates the Optimum Neuron container URIs to their latest versions and introduces a utility script to list published ECR images. The version is also bumped from 0.4.4.dev2 to 0.4.5.dev0.
Changes:
- Added a new Python script
tools/list-ecr-images.pyto programmatically list and retrieve container image tags from AWS ECR - Updated container URIs in documentation to reflect newer versions (Optimum Neuron 0.4.1 and 0.4.2, PyTorch 2.8.0, Transformers 4.55.4, vLLM 0.11.0, SDK 2.26.0)
- Incremented development version from 0.4.4.dev2 to 0.4.5.dev0
Reviewed changes
Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.
| File | Description |
|---|---|
| tools/list-ecr-images.py | New utility script for listing ECR images from HuggingFace Neuron repositories with AWS authentication and filtering |
| optimum/neuron/version.py | Version bump to 0.4.5.dev0 |
| docs/source/containers.mdx | Updated container URIs table with latest versions and added reference comment for the new listing script |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
tools/list-ecr-images.py
Outdated
| print(f"Warning: Docker login encountered an issue: {e}") | ||
|
|
||
|
|
||
| def get_ecr_images(ecr_client, repository_name: str, region: str, limit: Optional[int] = None) -> list: |
There was a problem hiding this comment.
The parameter 'region' is passed to get_ecr_images but never used within the function. This parameter should either be removed from the function signature or used if it's needed for future functionality.
|
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. |
Co-authored-by: Copilot <[email protected]>
Co-authored-by: Copilot <[email protected]>
Co-authored-by: Copilot <[email protected]>
Co-authored-by: Copilot <[email protected]>
| | Training | 0.4.1 | `763104351884.dkr.ecr.us-west-2.amazonaws.com/huggingface-pytorch-training-neuronx:2.8.0-transformers4.55.4-neuronx-py310-sdk2.26.0-ubuntu22.04` | | ||
| | Inference | 0.4.1 | `763104351884.dkr.ecr.us-west-2.amazonaws.com/huggingface-pytorch-inference-neuronx:2.8.0-transformers4.55.4-neuronx-py310-sdk2.26.0-ubuntu22.04` | | ||
| | vLLM | 0.4.2 | `763104351884.dkr.ecr.us-west-2.amazonaws.com/huggingface-vllm-inference-neuronx:0.11.0-optimum0.4.2-neuronx-py310-sdk2.26.0-ubuntu22.04 | | ||
|
|
There was a problem hiding this comment.
maybe we could consider generating this bit when producing the documentation, using the new tool?
What does this PR do?
This updates the URIs of the latest containers (and also adds a small script to list published containers).