Course Description
This IIoT for Machine Learning lecture series gives examples, walks through code, and focuses heavily on machine learning.
What am I going to get from this course?
Think of an Industrial Internet of Things application which will benefit their organization of employment, know how to obtain the data via real time data exchange and telecommunications, understand the code involved, remember the importance of data preparation from the prerequisite class, be introduced to algorithm building, automate the algorithm(s), and build and deploy the predictive system.
Prerequisites and Target Audience
What will students need to know or do before starting this course?
The prerequisite course is the Robot Applications for Machine Learning course because that course covers how to select technology and data preparation topics in much more detail. This course covers neural networks in greater detail.
Who should take this course? Who should not?
This course is aimed for implementers, but is helpful for executives who need to know what is involved for budgetary and scheduling purposes. Executives can choose to skip the code walk throughs.
Curriculum
Module 1: Introduction
05:38
Who your instructors are.
Lecture 2
Entire course outline
05:02
Top level review of what is covered in 29 lectures.
Module 2: Application Examples.
06:20
Lecture 3
Applications
06:20
Chemical, paint, and oil industry applications are covered.
Module 3: Real Time Data Exchange
31:42
Lecture 4
Real Time Data Exchange
00:53
Topics are sensors with WiFi
https://drive.google.com/drive/u/0/folders/0B-cBQwBR1spfUm9KTWJZNjV5TWM M2MIV.avi
Lecture 5
Real Time Data Exchange
23:12
Actual code walk through.
Lecture 6
Program Platform
02:07
Best choices for hardware and software.
Lecture 7
Sample code
02:14
Refers to the code walk through in lecture 5 of this module.
Lecture 8
Sample Comma Separated Variable File
01:49
The power is in the details.
Lecture 9
Sample Variable List
00:36
This will change for any specific application.
Lecture 10
Sample Application 1
00:17
Sample frequency and variables are presented.
Lecture 11
Sample Application 2
00:34
Frequency of sampling and determination of predictive variables are covered.
Start thinking further about your application.
Module 4: Telecommunication
02:41
Lecture 12
Best practices
02:03
Best practices, do's and don't's, and code walk through.
Lecture 13
Telecommunication Do's and Don't's
00:38
Avoid the following pitfalls.
Walk through of code which becomes firmware on board for controlling telecommunication.
Module 5: Machine Learning
Resource 2
Data Preparation
Follow these steps to giving your predictive model(s) a good basis for learning.
Resource 3
Building Algorithm
Tips for saving time during machine learning for initial model and also for implementation for deployment and installation.
Resource 5
Machine Learning
At least 10 different neural network functions are presented along with some code examples/
Resource 6
Six Sigma Alerts
These alerts are file driven by user so that each user can determine based on their responsibilities, if they want amber, orange, or red alerts or some combination thereof.
The log files are always written because they provide a clue as to what is skewing in the model or the hardware usage.
Like logging, the reports are always written to a file which can be accessed with the proper codes for information based on either timing and/or alerts.
Documentation is of two type, one for the systems administrator and the other for the user.
Training strategies are discussed.
Deployment strategies are discussed.
A system that is not getting requests for updates is not being used.
Module 7: Course Evaluation
Resource 13
Feedback to Instructor Form
Resource 14
Instructor Contact Information
Where to send the evaluation form.
