What does this Amazon Web Services Solution do?
Machine learning bot is an out-of-the-box solution that helps you quickly apply machine learning to actual business use cases in a simple and direct way. This solution based on Amazon SageMaker, turns the creation and management of machine learning models into an automated process. The solution allows you to quickly utilize machine learning without a professional background in artificial intelligence or machine learning.
This solution offers a graphic Web UI. Through the UI, you can upload image or text data of your specific application scenarios, and quickly complete the entire process from data labeling, model training to model evaluation. All processes will be completed in your own private Amazon Web Services account, so you don't need to worry about data security.
In addition, this solution allows you to freely use machine learning models. For the trained model, you can choose to automatically deploy it on the Amazon SageMaker endpoint for inference in the Amazon Web Services Cloud, or you can export it and apply it to scenarios at edges (such as factories, mobile client, data centers, etc.).
Currently, this solution supports three machine learning scenarios: Image Classification, Object Detection, and Name Entity Recognition.
Amazon Web Services Solution overview
The diagram below shows the architecture of this solution. You can contact us to obtain the implementation guide and Amazon CloudFormation deployment template.
This solution uses Amazon CloudFront to provide end users with a graphical Web UI. Users can use the machine learning functions provided by this solution on any device with a browser installed. Resources related to the Web UI are stored in an Amazon S3 bucket deployed with the solution.
This solution deploys two Amazon EC2 Auto Scaling Groups, which are used for "training tasks" and "inference tasks" respectively. After the training task is created, the "training tasks" auto scaling group will read the training data from the Amazon S3 training data bucket, or directly capture the training images through the camera through the Web UI, and store the model in the Amazon S3 model storage bucket after the training is completed. When the model is stored, the
"inference tasks" auto scaling group will receive an Amazon EventBridge notification, and the inference model will be automatically deployed to the group. Users can make calls to the endpoint to trigger the machine learning inference directly.
Graphic user interface
Easy to use
Browse our portfolio of Amazon Web Services-built solutions to common architectural problems.
Find Amazon Web Services certified consulting and technology partners to help you get started.
Sign-up and start exploring our services.