Amazon SageMaker 1O1
Amazon SageMaker is a fully managed machine learning service. Amazon SageMaker provides every developer and data scientist with the ability to build, train & deploy machine learning models quickly. It provides an integrated Jupyter authoring notebook instance for easy access to data sources for exploration and analysis, so users don’t have to manage servers. It provides common machine learning algorithms that are optimized to run efficiently against extremely large data in a distributed environment.
Amazon SageMaker includes three modules: Build, Train & Deploy. The Build module provides a hosted environment to work with user’s data, experiment with algorithms & visualize their output. The Train module allows for one-click model training and tuning at high-scale and low cost. The Deploy module provides a managed environment for users to easily host and test models for inference securely and with low latency.
Build:
Build highly accurate training data sets
Managed Notebooks for Authoring Models
Built-in, High Performance Algorithms
Broad Framework Support
Reinforcement Learning Support with Amazon SageMaker RL
Test and Prototype Locally
Train:
One-click Training
Managed Spot Training
Automatic Model Tuning
Train Once, Run Anywhere
Model tracking capability
Deploy:
One-click Deployment
Fully-managed Hosting with Auto Scaling
Batch Transform
Inference Pipelines
With native support for bring-your-own-algorithms and frameworks, Amazon SageMaker offers flexible distributed training options that adjust to users specific workflows. This is a HIPAA Eligible Service. To train a model in Amazon SageMaker, users need to create a training job. The training job includes the following information:
- The URL of the S3 bucket where users have stored the training data.
- The compute resources that users want SageMaker to use for model training.
- The URL of the S3 bucket where users want to store the output of the job.
- The Amazon Elastic Container Registry path where the training code is stored.
User can deploy model to get predictions in one of two ways:
- To set up a persistent endpoint to get one prediction at a time, use Amazon SageMaker hosting services.
- To get predictions for an entire dataset, use Amazon SageMaker batch transform.
Amazon SageMaker provides an Apache Spark library, in both Python and Scala, that users can use to easily train models in Amazon SageMaker using org.apache.spark.sql.DataFrame data frames in their Spark clusters. The Amazon SageMaker Python SDK TensorFlow estimators and models and the Amazon SageMaker open-source TensorFlow containers make writing a TensorFlow script and running it in Amazon SageMaker easier. Users can use SageMaker to train and deploy a model using a custom MXNet code.
Amazon SageMaker conforms to the AWS shared responsibility model, which includes regulations and guidelines for data protection. Users can use encrypted S3 buckets for model artifacts and data, as well as pass an AWS KMS key to SageMaker notebooks, training jobs, hyperparameter tuning jobs, batch transform jobs & endpoints, to encrypt the attached machine learning storage volume. Batch and training job containers and their storage are ephemeral in nature. When the job completes, the output is uploaded to S3 and the instance is torn down. The Amazon SageMaker folder in the ML Amazon EBS volume is the default storage location when the user opens a notebook instance. Amazon SageMaker saves any files within the SageMaker folder. SageMaker ensures that machine learning model artifacts and other system artifacts are encrypted in transit and at rest. Requests to the SageMaker API and console are made over a secure SSL connection. FIPS (Federal Information Processing Standard) validated endpoints are available for the SageMaker API and request a router for hosted models (runtime).
Next Steps: Read about
Amazon SageMaker RL https://docs.aws.amazon.com/sagemaker/latest/dg/reinforcement-learning.html
Amazon SageMaker Ground truth https://docs.aws.amazon.com/sagemaker/latest/dg/sms.html
Stay tuned for more info on ML and SageMaker. Reach out to me for any query at piyush.jalan93@gmail.com.
Thanks
Subscribe to FAUN topics and get your weekly curated email of the must-read tech stories, news, and tutorials 🗞️
Follow us on Twitter 🐦 and Facebook 👥 and Instagram 📷 and join our Facebook and Linkedin Groups 💬
