This repository contains two sophisticated search applications: Image Search and Document Search. Both applications are designed using AWS technologies and advanced machine learning models. They provide intuitive UIs for user interaction and leverage AWS Amplify, AppSync, and Lambda for backend processing.
- AWS Amplify with Angular
- AWS AppSync
- AWS Lambda
- AWS S3
- AWS Textract (Document Search)
- MongoDB
- Machine Learning Models (CLIP, Amazon Titan, Anthropic Claude)
- Image-search: Contains all code and resources for the Image Search Application.
- Document-search: Contains all code and resources for the Document Search Application.
- Upload and process images to search through a MongoDB vector database.
- Uses the CLIP model for image processing.
- Angular-based Amplify UI for image upload and search results display.
- Upload documents and perform question-answering based on content.
- Utilizes AWS Textract for document processing and Amazon Titan, Anthropic Claude models for question-answering.
- Amplify UI features a ChatGPT-like interface for user queries.
- AWS account
- Amplify CLI installed
- Node.js installed
- MongoDB
-
Clone the Repository
gh repo clone mongodb-partners/AppModernization_Amplify_AppSync_with_MongoDB_Atlas_Vector_Search -
AWS Configuration
- Configure AWS services like S3, Lambda, Textract, and AppSync following AWS documentation.
-
Amplify CLI Setup
- Setup Amplify CLI following the documentation
- Navigate to the Image Search UI.
- Upload images and view search results based on image content.
- Navigate to the Document Search UI.
- Upload documents, ask questions, and receive answers based on the document content.