Soheel Chughtai
IBM Early Experience Program Manager (eXp) - Watson Developer Cloud, IBM Systems
- Course NumberDW0101EN
- Classes StartFeb. 13, 2019
- Estimated Effort4 hours
- Audience
- Course Level
- Language
- Badge Earned
- Tell Your Friends
About this course
If you're a developer and want to learn about machine learning, this is the course for you. Even if you have some experience with machine learning, you might not have worked with audio files as your source data. Either way, you've come to right place.
In this course, you'll learn to create basic machine learning models that you train to recognize the sounds of dogs, cats, and birds. You'll also integrate visual recognition to identify images of these animals. You'll build a basic user interface in Node-RED that shows the results of the predictions for both sound and images.
You'll use IBM Watson Studio to build classification models to predict and identify animal sounds and use IBM Watson Visual Recognition to identify images of those animals. You'll learn how best to gather and prepare data, create and deploy models, deploy and test a signal processing application, create models with binary and multiclass classifications, and display the predictions on a web page.
Course syllabus
- IBM Watson Studio
- Lab 1: Gather and prepare the data
- Lab 1 overview
- Identify a source data set
- Lab 1 quiz
- Lab 2: Build a machine learning model
- Lab 2 overview
- 1. Create a machine learning project
- 2. Create a machine learning model
- 3. Deploy the machine learning model
- Lab 2 quiz
- Lab 3: Create predictions in a Node-RED application
- Lab 3 overview
- 1. Predict the animal sound
- 2. Deploy and test the signal processing application
- 3. Run predictions against the audio files
- Lab 3 quiz
- Lab 4: Create multiclass classification models
- Lab 4 overview
- 1. Create the first birdsong classification model
- 2. Create the remaining birdsong classification models
- 3. Make birdsong predictions from Node-RED
- Lab 4 quiz
- IBM Watson Visual Recognition
- Lab 5: Create UIs and integrate visual recognition
- Lab 5 overview
- 1. Create the interface for the dog and cat audio
- 2. Add the Visual Recognition service
- 3. Create the interface for the bird song audio
- Lab 5 quiz
- Final exam
Prerequisites
This course is for developers who have little or no experience with machine learning. No data science background required. You must know how to work with Node-RED.
You need the following accounts:
-
Kaggle (to be able to download a data set of audio files)
Kaggle is dedicated to data science and machine learning and hosts data sets that can be used to generate machine learning models. The source audio files that you will use in this course is from Kaggle.
You need the following software:
- Python 3
- Git command line
- PIP
- A suitable IDE to modify Python code, for example, Atom or Sublime
Course instructors
Emma Dawson
Emerging Technology Specialist, IBM Research
Course staff
Michelle Carey
IBM Courseware Developer
Frequently asked questions
Do I need experience in machine learning?
No, but it's helpful if you understand object-oriented programming and have some experience with Node-RED.
What web browser should I use?
The Open edX platform works best with current versions of Chrome, Firefox or Safari, or with Internet Explorer version 9 and above.
See our list of supported browsers for the most up-to-date information.