|
Today, we’re announcing a new feature in Amazon SageMaker Studio that simplifies and accelerates the machine learning (ML) development lifecycle. Amazon Q Developer in SageMaker Studio is an artificial intelligence generative assistant built natively into the SageMaker JupyterLab environment. This assistant takes input from your natural language and creates a tailored execution plan for your ML development lifecycle by recommending the best tools for each task, providing step-by-step instructions, generating code to get started, and offering troubleshooting assistance when you encounter mistakes. It also helps with problem solving, such as translating complex ML problems into smaller tasks and finding relevant information in the documentation.
You may be a first-time user evaluating Amazon SageMaker for generative artificial intelligence (Generative AI) or traditional ML use cases, or a returning user who knows how to use SageMaker but wants to further increase productivity and speed time to insights. With Amazon Q Developer in SageMaker Studio, you can build, train, and deploy ML models without leaving SageMaker Studio to search for sample notebooks, code snippets, and instructions on documentation pages and online forums.
Now let me show you the different options of Amazon Q Developer in SageMaker Studio.
Getting started with Amazon Q Developer in SageMaker Studio
In the Amazon SageMaker console, go to domains under Admin configuration and enable Amazon Q Developer in the domain settings. If you are new to Amazon SageMaker, see the Amazon SageMaker domain overview documentation. I am choosing Studio of Commencement drop-down list mytestuser to launch Amazon SageMaker Studio.
When my environment is ready, I choose JupyterLab under Application and then choose Open JupyterLab open your Jupyter notebook.
The Amazon Q Developer generative assistant with artificial intelligence is next to my Jupyter notebook. There are built-in commands that I can now use to get started.
I can immediately start a conversation with an Amazon Q Developer by describing an ML problem in natural language. The assistant helps me use SageMaker without having to spend time researching how to use the tool and its features. I use the following prompt:
I have data in my S3 bucket. I want to use that data and train an XGBoost algorithm for prediction. Can you list down the steps with sample code.
Amazon Q Developer gives me step-by-step instructions and generates code to train the XGBoost algorithm for prediction. I can follow the recommended steps and easily add the desired cells to my notebook.
Let me try another challenge to generate code to download a dataset from S3 and load it using Pandas. I can use it to build or train my model. This helps to streamline the coding process by solving repetitive tasks and reducing manual work. I use the following prompt:
Can you write the code to download a dataset from S3 and read it using Pandas?
I can also ask Amazon Q Developer for debugging and bug fixing guidance. Assistant helps me troubleshoot based on common errors and solutions, saving me time-consuming online research and trial and error. I use the following prompt:
How can I resolve the error "Unable to infer schema for JSON. It must be specified manually." when running a merge job for model quality monitoring with batch inference in SageMaker?
As a final example, I ask Amazon Q Developer to provide me with recommendations on how to schedule a laptop job. I use the following prompt to get the answer:
What are the options to schedule a notebook job?
Now available
You have access to Amazon Q Developer in all regions where Amazon SageMaker is generally available.
The Assistant is available to all Amazon Q Developer Pro Tier users. For pricing information, see the Amazon Q developer pricing page.
Get started with Amazon Q Developer in SageMaker Studio today and access a generative AI assistant at any point in your ML development lifecycle.
-Esra
