Demonstrate running the HuggingFace NLP training script locally in the 'SageMaker' mode example notebook, to illustrate the workflow for debugging the script before running in SageMaker.
Issue #, if available: #33 (partial)
Description of changes:
Adjust the Hugging Face script mode example to support running the train.py locally on the notebook via !python command, and add the example command to the SageMaker mode notebook.
The code is commented out to avoid running by default, because the nb will run out of memory unless the previous 'Local Mode' kernel is killed: Training BORT is too memory-intensive to have 2 copies open on a t3.medium!
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.
Demonstrate running the HuggingFace NLP training script locally in the 'SageMaker' mode example notebook, to illustrate the workflow for debugging the script before running in SageMaker.
Issue #, if available: #33 (partial)
Description of changes:
Adjust the Hugging Face script mode example to support running the
train.pylocally on the notebook via!pythoncommand, and add the example command to the SageMaker mode notebook.The code is commented out to avoid running by default, because the nb will run out of memory unless the previous 'Local Mode' kernel is killed: Training BORT is too memory-intensive to have 2 copies open on a t3.medium!
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.