batermj / aws-ml-vision-end2end

Content consists of Jupyter Notebook tutorials walking through deep learning Frameworks (MXNet, Gluon) to Platforms (SageMaker, DeepLens) for common CV use-cases

AWS Ml Vision End2end

This repository contains Jupyter Notebook tutorials for computer vision use-cases. The tutorials take you end-to-end through the process of developing a deep-learning model for computer vision:

• Load, Explore, and Understand the data relevant to your computer vision task
• Prototype deep learning models in the MXNet Framework, using both the MXNet Symbolic API and the imperative Gluon interface.
• Port prototype code to run scalable training jobs using the Amazon SageMaker platform
• Deploy trained models to inference endpoints using Amazon SageMaker
• Deploy trained models to the Edge using AWS DeepLens.

Tutorials Available

• Image Classification: A level-101 Intro to Computer Vision with Deep Learning, this tutorial covers a very simple image classification task for traffic sign classification.

Coming Soon

• Transfer Learning
• Object-Detection
• Semantic Segmentation

Data Sets Used

• Traffic Sign Dataset

License

This library is licensed under the Apache 2.0 License.