Parameter description update #3777
Description
What did you find confusing? Please describe.
I tried to read the description for sagemaker.transformer.Transformer.transform()
https://sagemaker.readthedocs.io/en/stable/api/utility/inputs.html#sagemaker.inputs.TransformInput
Describe how documentation can be improved
Add parameter descriptions
Additional context
Here are some resources:
- readthedocs: https://sagemaker.readthedocs.io/en/stable/api/utility/inputs.html#sagemaker.inputs.TransformInput
- APIRef: https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TransformInput.html
- GitHub: https://github.com/aws/sagemaker-python-sdk/blob/e917cb47c51041e168a09c472b19e07fdfeaf1af/doc/api/utility/inputs.rst
I wrote an attempted draft:
@attr.s
class TransformInput(object):
"""Create a class containing all the parameters.
It can be used when calling ``sagemaker.transformer.Transformer.transform()``
Parameters:
data (str): The S3 location of the input data that the model can consume.
data_type (str): elastic inference accelerator type.
(default: ``'S3Prefix'``)
content_type (str): The multipurpose internet mail extension (MIME) type of the data.
(default: None)
compression_type (str): If your transform data is compressed, specify the compression type.
Valid values: ``'Gzip'``, ``None``
(default: None)
split_type (str): The method to use to split the transform job's data files into smaller
batches.
Valid values: ``'Line'``, ``RecordIO``, ``'TFRecord'``, None
(default: None)
input_filter (str):
(default: None)
output_filter (str):
(default: None)
join_source (str):
(default: None)
model_client_config (dict):
(default: None)
batch_data_capture_config (dict): Specifies configuration related to batch transform job
data capture for use with Amazon SageMaker Model Monitoring.
(default: None)
"""
data: str = attr.ib()
data_type: str = attr.ib(default="S3Prefix")
content_type: str = attr.ib(default=None)
compression_type: str = attr.ib(default=None)
split_type: str = attr.ib(default=None)
input_filter: str = attr.ib(default=None)
output_filter: str = attr.ib(default=None)
join_source: str = attr.ib(default=None)
model_client_config: dict = attr.ib(default=None)
batch_data_capture_config: dict = attr.ib(default=None)