AI for Time Series Analysis and Forecasting
Overview
Today, there is a great need for the introduction of AI into all aspects of the software, making the enterprise software smart. The argument one often finds in articles describing the unsatisfactory state of business software is, “If smartphones can do it, why can’t enterprise software?”
This course addresses the need for smart software for time series analysis and forecasting.
The course is intended for software architects and engineers. It gives them a practical level of experience, achieved through a combination of about 50% lecture, 50% demo work with student’s participation.
Duration
3 Days
Audience
Software Architects, Developers
Prerequisites
- familiarity with any programming language
- be able to navigate Linux command line
- basic knowledge of command-line Linux editors (VI / nano)
Lab environment
The working environment will be provided for students. Students would only need an SSH client and a browse.
Zero Install: There is no need to install software on students’ machines.
Course Outline
AI overview
-
- A brief history of AI
- Types of AI systems
- Training machine learning models
- Applying models for prediction
- Demos and Labs
Time series processing and forecasting elements
-
- Traditional Time Series forecasting with ARIMA models
- Defining Autocorrelation
- Understanding the Dickey-Fuller Test
Forecasting with TensorFlow and Keras
-
- Google democratization of AI with TensorFlow
- Types of neural network (Perceptron, CNN, LSTM) and their use
- Forecasting with TensorFlow
- Using RNN and LSTM in time series prediction.
- Validation and metrics of Time Series Prediction models
- Use cases and labs