Training on implementing Time series prediction with Deep Learning on Sagemaker and Tensorflow in AWS

I have time series data that I would like to run though a Tensorflow Deep Learning (RNN) on a VM in AWS with Jupyter. I have the bulk of the script written. Just need some tuning and training to QA the final result. Would like walkthrough in Sagemaker and also with dedicated Tensorflow machine to get a good feel for the infrastructure and how it all hangs together.