Training and deploying a Semantic Segmentation model using Amazon Sagemaker using the semantic segmentation algorithm from Sagemaker. In this project, we have trained and deployed a model that will be able to segment images of dogs and cats from the popular IIIT-Oxford Pets Dataset into 3 unique pixel values. That is, each pixel of an input image would be classified as either foreground (pet), background (not a pet), or unclassified (transition between foreground and background).
The code is derived from the Semantic Segmentation with Amazon Sagemaker project course on Coursera and this repository is created to learn and practice deploying Machine Learning models on AWS Sagemaker