sagemaker-python-sdk
sagemaker-python-sdk copied to clipboard
aws
A library for training and deploying machine learning models on Amazon SageMaker
**What did you find confusing? Please describe.** `py_version` argument documentation is outdated for estimator. e.g.: https://sagemaker.readthedocs.io/en/stable/frameworks/pytorch/sagemaker.pytorch.html?highlight=pytorch#pytorch-estimator It says py version should be one of ‘py2’ or ‘py3’. Both options are...
When I am experimenting the batch transform for a sklearn model, I observe an error when there is only 1 row of data. I suppose the batch transform should also...
*Issue #, if available:* *Description of changes:* audit doc string and function and constructor args type mismatch. *Testing done:* ## Merge Checklist _Put an `x` in the boxes that apply....
*Issue #, if available:* closes [#2527](https://github.com/aws/sagemaker-python-sdk/issues/2647) This is a clone of https://github.com/aws/sagemaker-python-sdk/pull/2644 but has allow commits from maintainers on so should be easier to merge. Description of changes: Add try...
**Describe the bug** Create a model without specifying sagemaker_session and you get this error **To reproduce** ``` model = Model( image_uri = "246618743249.dkr.ecr.us-west-2.amazonaws.com/sagemaker-scikit-learn:0.23-1-cpu-py3", model_data = my_model_data_url, role = role_arn, source_dir...
*Description of changes:* Currently the signature type `instance_type` and `instance_count` of the `PySparkProcessor.__init__` is wrong. https://github.com/aws/sagemaker-python-sdk/blob/284ddbebcf6240f0a4d3c734244f8e8ad066a9b3/src/sagemaker/spark/processing.py#L683-L684 *Testing done:* ## Merge Checklist _Put an `x` in the boxes that apply. You...
I am trying to import a model for use with torcheia. I am using the following code: ```python import torch def model_fn(): model = torch.jit.load("cos_1.5.1.pt", map_location=torch.device('cpu')) if torch.__version__ == '1.5.1':...
I want ro pass a list as a hyperparameter to an estimator. For example, hyperparameter key = 'a' hyperparameter value = ['def', 'xyz'] This type of value is accepted and...
**Describe the bug** The following code results in an exception: ```python fg = FeatureGroup( name="hello", sagemaker_session=feature_store_session ) fg.put_record([ FeatureValue('id', '2'), FeatureValue('data', 'test\n'), FeatureValue('EventTime', '0.0') ]) ``` The exception: ``` ClientError:...
