IBM Watson AutoAI machine learning tutorial | Running AutoAI
This is the second video in a three-part series from @Horea Porutiu.
Part 1: https://youtu.be/knxbJgPmD5E
Part 3: https://youtu.be/su79BFQ9VU0
Watson Machine Learning web-app demo using AutoAI: https://youtu.be/B1WAOiO1o4o
This installment shows you how to run AutoAI. In this video, we create the IBM Cloud services needed to run and automate the deployment of our machine learning models. Before you get started, you must have an IBM Cloud account, which you can register for in the link below. Next, you will create an IBM Watson Studio service, and upload your data set into your project. Next, you will create an AutoAI experiment, and add a Watson Machine Learning instance to your project, so that you can generate multiple machine learning models. Lastly, you will configure and run your AutoAI experiment, and produce 8 different pipelines that are ranked based on RMSE (root mean squared error). Lastly, we save our best performing pipeline as a model.
For more, visit these links:
- Free IBM Cloud account needed, register here: https://tinyurl.com/y4mzxow5
- The data set: https://www.kaggle.com/noordeen/insurance-premium-prediction
- Project GitHub: https://github.com/IBM/predict-insurance-charges-with-ai
If you want to see more of Horea’s content, focused on open-source, blockchain, and artificial intelligence, see his channel at: https://www.youtube.com/channel/UCSQpzlNDZOUR3LAaGWNMovw
Build Smart. Build secure. Join a global community of developers at http://ibm.biz/IBMdeveloperYT
#MachineLearning
#WebApplication
#AutoAI