sagemaker-python-sdk icon indicating copy to clipboard operation
sagemaker-python-sdk copied to clipboard

A library for training and deploying machine learning models on Amazon SageMaker

Results 337 sagemaker-python-sdk issues
Sort by recently updated
recently updated
newest added

**Describe the bug** `_RepackModelStep` is a part of the `sagemaker.workflow.model_step.ModelStep`. Its purpose is to attach a `source_dir` to the plain `model.tar.gz`. As I understand, this is done using a training...

bug

*Issue #, if available:* *Description of changes:* ModelBuilder to fetch local schema when no SchemaBuilder present *Testing done:* * Unit tests * Integ tests * Notebook test ## Merge Checklist...

*Issue #, if available:* *Description of changes:* *Testing done:* ## Merge Checklist _Put an `x` in the boxes that apply. You can also fill these out after creating the PR....

*Issue #, if available:* *Description of changes:* * Basic test to verify User Agent propagation by adding test_user_agent.py * Test to ensure that notebook instance is injected * Test to...

*Issue #, if available:* As the sagemaker pysdk is installed in both local client side and job side, we've received issues reported relating to the inconsistent sagemaker versions. For example,...

*Issue #, if available:* https://docs.google.com/document/d/1OTXyZ-YpovntrzzBVBOBCnVl4Ijdl858s35UJFKf3dI/edit *Description of changes:* Use the model size and given instance to check if we can deploy. Flow: 1. If metadata tag is “text generation” go...

*Issue #, if available:* *Description of changes:* - Fix telemetry image uri option logic for ModelBuilder *Testing done:* ## Merge Checklist _Put an `x` in the boxes that apply. You...

https://github.com/awslabs/llm-hosting-container/blob/main/huggingface/pytorch/optimum/docker/0.0.18/Dockerfile *Issue #, if available:* *Description of changes:* *Testing done:* ## Merge Checklist _Put an `x` in the boxes that apply. You can also fill these out after creating the...

do-not-merge

*Issue #, if available:* N/A *Description of changes:* This is a follow up change of the previous PR https://github.com/aws/sagemaker-python-sdk/pull/4372. With it, function step can be fully supported w.r.t selective pipeline...

*Issue #, if available:* *Description of changes:* Adds support for `hub_name` field in the `JumpStartEstimator` class. Users can pass in a Hub Name or Arn to this class in order...