sagemaker-python-sdk
sagemaker-python-sdk copied to clipboard
aws
A library for training and deploying machine learning models on Amazon SageMaker
*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:* For python upgrade 3.10, add sm sklearn container 1.2.1.1 and sm xgboost container 1.7.1.1 *Description of changes:* Add sm xgboost container: 1.7-1-1, and sklearn container 1.2-1-1...
*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:* *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:* *Testing done:* ## Merge Checklist _Put an `x` in the boxes that apply. You can also fill these out after creating the PR....
**Describe the feature you'd like** Sagemaker local uses docker compose internal network to make containers communicate with each other. For some tools, such as Snowflake, some companies may require login...
**Describe the bug** The `hyperparameters` in `XGBoost` are incorrectly [json-dumped](https://github.com/aws/sagemaker-python-sdk/blob/5699edfe8ec121b468ecf3c8830a8de06c7f6c66/src/sagemaker/estimator.py#L921) with extra quotes which causes the validation to fail. **To reproduce** sample code ```python from sagemaker.xgboost.estimator import XGBoost params =...
**Describe the bug** So I've used the same code for FeatureGroup ingestion for years, and I noticed the other day that ingest() now just hangs. This seems to be correlated...
*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....
**Describe the feature you'd like** Right now, the `sagemaker-python-sdk` pins certain dependencies to extremely old and restrictive versions. The biggest offender is `numpy == 1.26.4` which is over 18 months...
