This repo contains materials for the "Modeling timeseries and sequence data on AWS using Apache MXNet and Gluon" workshop at Applied ML Days 2018. In this workshop we learn how to model generic sequence data and timeseries data to do sequence prediction, forecasting future timeseries, anomaly detection to name a few. We will use Apache MXNet a scalable deep learning framework to train our neural networks. Once trained we will explore various scalable deployment options on AWS for both real-time and batch inference.
refer to slides for an introduction in to time series modeling on AWS
- Introduction to ApacheMXNet and Gluon - Crash Course
- Introduction to TimeSeries Analysis
- Univariate TimeSeries
- Multivariate TimeSeries Forecasting
- TimeSeries Forecasting with DeepAR
- Seq2Seq with Apache MXNet
- Setup an AWS account
- Option 1: Run on Amazon SageMaker
- Option 2: Setup a notebook server on AWS DL AMI