Predictive applications continue to be at the forefront of many AWS service uses. AWS recently released a fascinating post where they highlight the need and technical process of applying predictive models and ML features into patient predictions within multiple healthcare organisations.
Not only are they able to document the deep learning model and the theory behind it’s creation but they are also able to highlight the importance of interpretation of results from the models using visualisation techniques.
Here at Firemind, we also understand that the interpretation and visualisation of data is often just as important as the data itself. Without a clear way to evaluate what you’re gathering ‘data wise’, you can be left with a vast amount of data and no actionable tasks!
The architecture diagram above illustrates the model training pipeline, inference pipeline and the front end rendering the information.
What You’ll Find in the AWS Post
If you visit the article, you’ll be able to see further information on:
1. Creating, training and testing datasets.
2. Using embedding techniques for a richer representation of the unstructured data.
3. Training the models.
4. Evaluating the results.
5. Visualising the results with custom UI components.
How can Firemind assist with an ML AWS Project?
Here at Firemind, we’ve worked with countless clients to integrate AI/ML AWS services into their businesses. From utilising Amazon SageMaker to build, train and deploy machine learning models to crafting Amazon Rekognition models to automate image and video analysis.
If you’d like to speak with us about an AI/ML Project, please Get in Touch today and a member of our Sales team would be happy to discuss your requirements.