sagemaker-python-sdk

sagemaker-python-sdk copied to clipboard

Add typing information for some processors

5

*Issue #, if available:* #3213 *Description of changes:* Add types to classes inheriting from FrameworkProcessor and the run method. *Testing done:* Have not tested, changes are only types. I am...

MatthewCaseres

Environment variables limited to 256 characters

4

**Describe the bug** Currently, the value length constraints for environment variables are quite limited in `CreateProcessingJob` and `CreateTrainingJob`. * [CreateProcessingJob](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateProcessingJob.html#sagemaker-CreateProcessingJob-request-Environment): 256 * [CreateTrainingJob](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingJob.html#sagemaker-CreateTrainingJob-request-Environment): 512 This causes issues when passing urls...

jmahlik

ConditionComparison fails if `left` is a primitive such as `float` or `int`

2

**Describe the bug** ConditionComparison fails if `left` is a primitive such as a `float` or `int`. https://github.com/aws/sagemaker-python-sdk/blob/0a0b6e65288496ad33fd5ee34020665aced02156/src/sagemaker/workflow/conditions.py#L82 This line should be `"LeftValue": primitive_or_expr(self.left)` matching the line below it or, if...

kevinnowland

Why the supported PySpark version (2.4) is so outdated given spark 3.2 is available?

2

Very simple request: is it possible to keep track the most recent major version of PySpark and support it in `PySparkProcessor`? If the PySpark is so outdated in `PySparkProcessor`, I...

jtzhang17

**Describe the feature you'd like** I want type validations when I am constructing FrameworkProcessors and calling their run method. **How would this feature be used? Please describe.** It will help...

MatthewCaseres

feature: Adding support for Multi Worker Mirrored Strategy in TF estimator

103

*Issue #, if available:* Resolves https://github.com/aws/sagemaker-tensorflow-training-toolkit/issues/391 *Description of changes:* SageMaker Tensorflow Training ToolKit now supports Multi Worker Mirrored Strategy through cluster initialization. These changes allow the cluster init to be...

Lokiiiiii

.predict_async(data) called on AsyncPredictor API fails if no name was provided to predictor

1

**Describe the bug** In the SageMaker SDK, if no name is given to an `AsyncPredictor`, and you call `.predict_async(data)` with `data` and not an `input_path`, the SDK errors out. **To...

joaopcm1996

Model.compile ignores framework_version when compiling for ml_inf1

1

**Describe the bug** When attempting to compile for ml_inf1 via the SDK a model which was trained / fine-tuned with PyTorch 1.9.1, the `framework_version` argument is ignored, resulting in a...

vprecup

PyTorch: increasing --shm-size to allow multiprocessing data loaders

17

Reference: MLFW-2582 ### System Information - **Framework**: PyTorch - **Framework Version**: 1.1.0 - **Python Version**:py3 - **CPU or GPU**: Both - **Python SDK Version**: 1.32.0 - **Are you using a...

ksanjeevan

Linking Readthedocs to Example Notebook Readthedocs

10

*Issue #, if available:* N/A *Description of changes:* A few changes: 1. Adds Jinja2==3.0.0 to requirements.txt file for it to be able to build, I am unable to build locally/read...

rdamazon