nerdimite / sentiment-analysis-deployment

End to End Sentiment Analysis Project (Udacity Machine Learning Engineer Nanodegree)

End-to-End Sentiment Analysis

This repository contains the Sentiment Analysis Deployment Project for Udacity Machine Learning Engineer Nanodegree. The Sagemaker_Project directory contains the jupyter notebook and python files of the model, training and prediction scripts.

Project Details

• The sentiment analysis is built using Pytorch and deployed in AWS Sagemaker as an Endpoint
• The deployed model Endpoint communicates to AWS Lambda which in turn communicates to AWS API Gateway which listens to REST POST request from the web app
• The web app sends a post request to API Gateway which triggers the Lambda function and invokes the model Endpoint which then returns the prediction as a response