Deploy a Production Machine Learning model with AWS & React

Build a Scalable and Secure, Deep Learning Image Classifier with SageMaker, Next.js, Node.js, MongoDB & DigitalOcean

Created by: Patrik Szepesi
Machine Learning Engineer

What Will I Learn?

• Deploy a production ready robust, scalable, secure Machine Learning application
• Set up Hyperparameter Tuning in AWS
• Find the best Hyperparameters with Bayesian search
• Use Matplotlib, Numpy, Pandas, Seaborn in SageMaker
• Use AutoScaling for our deployed Endpoints in AWS
• Use multi-instance GPU instance for training in AWS
• Learn how to use SageMaker Notebooks for any Machine Learning task in AWS
• Set up AWS API Gateway to deploy our model to the internet
• Secure AWS Endpoints with limited IP address access
• Use any custom dataset for training
• Set up IAM policies in AWS
• Set up Lambda concurrency in AWS
• Data Visualization in SageMaker
• Learn how to do MLOps in AWS
• Build and deploy a MongoDB, Express, Nodejs, React/nextjs application to DigitalOcean
• Create an end to end machine learning pipeline all the way from gathering data to deployment
• File Mode vs Pipe Mode when training deep learning models on AWS
• Use AWS' built in Image Classifier
• Create deep learning models with AWS SageMaker
• Learn how to access any AWS built in algorithm from AWS ECR
• Use CloudWatch logs to monitor training jobs and inferences
• Analyze machine learning models with Confusion matrix, F1 score, Recall, and Precision
• Access AWS endpoint through a deployed MERN web application running on DigitalOcean
• Build a beautiful web application
• Learn how to combine AI and Machine Learning with Healthcare
• Set up Data Augmentation in AWS
• Machine Learning with Python
• JavaScript to deploy MERN apps

Requirements

• Any laptop and an internet connection
• Some Python and Machine Learning Knowledge
• about 15-40 dollars for using AWS resources(Optional, only applies if you follow along with me)

Target audience

• Those with some ML experience who are hoping to take their skills to the next step by being able to deploy their deep learning models to production