In this repository we will go through the exercise of deplying a sentiment analysis model on AWS SageMaker. To to do we will use the IMDb dataset.
To determine the sentiment of reviews we use a recurrent neural network (RNN). The picture below shows a sample architecture.
The project begins with collecting the data, cleaning it in a readable format, and running some tests within the notebook. Next we deploy the model to a web app using AWS's Lambda. The sketch below shows the interaction between the app and the deployed model.
With our model deployed, we can access the URL associated with the endpoint. The plot below shows a sample review.