aws-step-functions-data-science-sdk-python
aws-step-functions-data-science-sdk-python copied to clipboard
aws
Step Functions Data Science SDK for building machine learning (ML) workflows and pipelines on AWS
When an event includes boolean properties the `execution.list_events(html=True)` method fails with error: ``` --------------------------------------------------------------------------- TypeError Traceback (most recent call last) in ----> 1 execution.list_events(html=True) # Bug ~/anaconda3/envs/python3/lib/python3.6/site-packages/stepfunctions/workflow/stepfunctions.py in list_events(self, max_items,...
When running notebooks in an automated fashion many times the StepFunction created by the workflow.create() method already exists. The same for when running the workflow.update() method when defining the Workflow...
The EndpointConfig constructor could be extended to support passing in AsyncInferenceConfig which includes configuration as shown here in the doc [API_AsyncInferenceConfig](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AsyncInferenceConfig.html) In this way would be possible to create an...
Currently, it is not possible to update the estimator hyperparameters with the hyperparameters passed to TrainingStep if a Placeholder is used as input. The merging of hyperparameters can only be...
While an example includes use of PipelineModel: https://github.com/aws/aws-step-functions-data-science-sdk-python/blob/main/src/stepfunctions/template/pipeline/inference.py#L128 This is a no-op: https://github.com/aws/aws-step-functions-data-science-sdk-python/blob/main/src/stepfunctions/template/pipeline/inference.py#L142 Setting the parameters property results in no changes to the CreateModel API call. Instead, this method should...
Currently I am not able to pass execution input to a Map state. If I try to pass execution input to map state it throws 'Object of type 'ExecutionInput' is...
Request: Add aws-step-functions-data-science-sdk-python to the community led conda-forge repository at https://github.com/conda-forge via their add recipes instructions from https://conda-forge.org/#add_recipe Consistent to SageMaker https://anaconda.org/conda-forge/sagemaker-python-sdk Example usage: `$ conda install -c conda-forge sagemaker-python-sdk`...
Hi, sagemaker requires packaging>=20 but scpetre requires 16.8 or below. Both have conflict when I try to use them together. `pkg_resources.VersionConflict: (packaging 20.9 (c:\users\c279782\appdata\local\programs\python\python36\lib\site-packages), Requirement.parse('packaging==16.8'` Is there a workaround ?...
Currently need to set the name manually for `StepId.ProcessingJob.value`, as it does not show-up in StepID. see code under *Setup Pipeline with the Step Functions SDK* for example: https://github.com/data-science-on-aws/workshop/blob/master/10_pipeline/stepfunctions/01_Create_Pipeline_Train_and_Deploy_Reviews_BERT_TensorFlow.ipynb
There is not StepID for TuningStep. Would expected `StepId.TuningJob.value` to yield the job name. Does not look like it's currently included.