-
Notifications
You must be signed in to change notification settings - Fork 1.2k
Description
Please fill out the form below.
System Information
- Python Version: 3.6
- Python SDK Version: 1.18
Describe the problem
I'm trying to fit with an Estimator on Gzipped data with Pipe mode. I don't have input_mode set on the Estimator, but I do have it set in the s3_input, which should override the Estimator's input_mode:
sagemaker-python-sdk/src/sagemaker/session.py
Lines 1329 to 1336 in c2bac8f
| input_mode (str): Optional override for this channel's input mode (default: None). By default, channels will | |
| use the input mode defined on ``sagemaker.estimator.EstimatorBase.input_mode``, but they will ignore | |
| that setting if this parameter is set. | |
| * None - Amazon SageMaker will use the input mode specified in the ``Estimator``. | |
| * 'File' - Amazon SageMaker copies the training dataset from the S3 location to a local directory. | |
| * 'Pipe' - Amazon SageMaker streams data directly from S3 to the container via a Unix-named pipe. | |
My s3_input is:
training_s3_input = s3_input('s3://my_training_data', compression='Gzip', input_mode='Pipe', shuffle_config=ShuffleConfig(1))
Trying to fit on an Estimator gives me back this ValidationError, even though I specify Pipe, not File:
An error occurred (ValidationException) when calling the CreateTrainingJob operation: Invalid compression type for channel training: File mode only supports NONE, got Gzip instead
Setting input_mode='Pipe' directly on the Estimator works as expected.