AWS Launches ‘SageMaker Studio Lab’: A Free Tool To Learn and Experiment With Machine Learning

AWS introduced SageMaker Studio Lab, a free offering to assist developers master machine learning techniques and experimenting with the technology, at its re: Invent 2021 Event. Users get everything they need to start with Studio Lab, including a JupyterLab IDE, model training on CPUs and GPUs, and 15 GB of persistent storage.

ML for everyone

As the use of machine learning expands, so does the demand for machine learning expertise. To satisfy this rising demand, AWS is expanding the reach of machine learning (ML) beyond data scientists and developers to include line-of-business analysts who assist finance, marketing, operations, and HR departments. AWS stated that Amazon SageMaker Canvas would increase access to machine learning by giving business analysts a visual point and click interface that allows users to produce accurate ML forecasts without writing a single line of code. Get started with a two-month free trial that includes up to ten machine learning models and up to one million cells of data.

At scale, processing structured and unstructured data

As more employees begin to use machine learning in their daily jobs, the requirement to label datasets for training rises, and data science teams cannot keep up. Amazon SageMaker Ground Truth Plus was released by AWS to make it simple to construct high-quality training datasets without needing to build labeling apps or maintain labeling workforces yourself. SageMaker Ground Truth Plus can help you satisfy your data security, privacy, and compliance standards by providing an experienced crew that has been educated in machine learning activities. Simply submit your data, and Amazon SageMaker Ground Truth Plus will design and maintain data labeling workflows for you. To get started with it, submit a request for a pilot.

Optimize ML model development, training, and deployment performance and cost.

AWS is also making it easier and less expensive for data scientists and developers to prepare data and construct, train, and deploy machine learning models.

First, AWS improved Amazon SageMaker Studio for constructing machine learning models, allowing users to execute data processing, analytics, and machine learning processes all in one place. You may access a wide range of data sources and write code for any transformation for various data tasks from this universal notebook.

AWS also released a new compiler, Amazon SageMaker Training Compiler, which can speed up training by up to 50% using graph and kernel-level optimizations to make GPUs more efficient. SageMaker Training Compiler is compatible with TensorFlow and PyTorch versions in SageMaker. As a result, you can accelerate training in these popular frameworks with a few code modifications.

Finally, AWS introduced two capabilities for an inference that would lower inference costs. Amazon SageMaker Serverless Inference (preview) allows you to install machine learning models on a pay-per-use basis without worrying about clusters or servers for use cases with irregular traffic patterns. Furthermore, Amazon SageMaker Inference Recommender assists you in selecting the best available compute instance and configuration for ML model deployment for optimal inference performance and cost.

Free ML training

Amazon SageMaker Studio Lab (preview) is a free machine learning notebook environment that allows anybody to experiment with developing and training machine learning models without worrying about infrastructure or identity and access management. SageMaker Studio Lab speeds up model development by integrating with GitHub. It comes preconfigured with the most popular machine learning tools, frameworks, and libraries to quickly get you up and running. SageMaker Studio Lab provides 15 GB of dedicated storage for your machine learning projects, and it saves your projects automatically, so you don’t have to restart between sessions. It’s as simple as shutting off your computer and returning later. All you need to get started with SageMaker Studio Lab is a valid email address.

Tool: https://studiolab.sagemaker.aws/

References

• https://aws.amazon.com/blogs/machine-learning/roundup-of-reinvent-2021-amazon-sagemaker-announcements/
• https://techcrunch.com/2021/12/01/aws-launches-sagemaker-studio-lab-a-free-tool-for-learning-machine-learning/

Suggested