# 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](https://nbviewer.jupyter.org/github/aws-samples/aws-ml-vision-end2end/blob/master/Image_Classification/Image_Classification_Tutorial.ipynb): 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](http://benchmark.ini.rub.de/?section=gtsrb&subsection=dataset#Imageformat) 

## License

This library is licensed under the Apache 2.0 License.