From the course: Machine Learning with SageMaker by Pearson

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Setting up and running SageMaker training jobs

Setting up and running SageMaker training jobs - Amazon SageMaker Tutorial

From the course: Machine Learning with SageMaker by Pearson

Start my 1-month free trial Buy for my team

Setting up and running SageMaker training jobs

“

Once we have our data set clean and ready to go, it's time to train a model. And we do that by creating a SageMaker training job. This is a process for training your machine learning models. This does use Elastic Compute Cloud instances for the compute. So there are instance types such as T2.micro. T2 is an instance generation, and then micro is the size. And the size equates to the number of virtual CPUs and the amount of memory that's available to your instance. And we'll talk about different instance types here in just a moment. There is support for distributed training, as well as multi-GPU training. So you could have 10 instances of GPU-enabled instances and distribute your training job across all 10. The automation is there for scaling, as well as infrastructure setup. So you define your script and its dependencies, upload that data into S3, specify the training container and the algorithms that you're going to use, configure your hyperparameters for the model. Recall from a…

Contents