aws-step-functions-data-science-sdk-python icon indicating copy to clipboard operation
aws-step-functions-data-science-sdk-python copied to clipboard

Published 20 hours ago • verified publisher icon aws

Add `deploy_instance_count` and `deploy_instance_type` to `TrainingPipeline`

Open cfregly opened this issue 5 years ago • 2 comments
trafficstars

Currently, TrainingPipeline uses the same instance type and count for both train and deploy.

Different instance types and counts are desirable to address the different profiles for each workload.

cfregly avatar Jun 16 '20 04:06 cfregly

Thanks for the feedback @cfregly. The use case of having different instance types and counts for training and the endpoint is a valid one. We will look into this.

vaib-amz avatar Jul 24 '20 22:07 vaib-amz

Cool. we ended up hacking it here: https://github.com/data-science-on-aws/workshop/blob/a7c5603/10_pipeline/01_Create_Pipeline_Train_and_Deploy_Reviews_BERT_TensorFlow.ipynb

(search for TrainingPipelineWithDifferentDeployInstanceType)

cfregly avatar Jul 24 '20 23:07 cfregly