Written by: @Jeewoo Chung
Introduction
Hey everyone! Thanks again for coming to today’s meeting. I hope you had fun and learned something that you found cool today. Machine Learning is something that has been a hot topic for a while now and is definitely an industry that you could go into if you found it very interesting — for those of you who weren't able to join us today, or for those who want to review what we did during our club meeting, here's a recap document that you can duplicate to your Notion workspace, as an archive.
Today's Meeting Recap
Hack Club Workshop Link: https://workshops.hackclub.com/teachable_machine/
Teachable Machine by Google: https://teachablemachine.withgoogle.com/
Machine Learning
What is Machine Learning?
- The study of computers that improve itself automatically through experience (trial and error)
- It is seen as a part of artificial intelligence (AI)
Machine Learning Model
- A mathematical model for the process of machine learning
- These "models" are trained with large amounts of data that attempt to teach the model
- The more data there is the more it's going to be able to "learn"
→ i.e. Difference between two sets of data: the more examples there are to compare, the more accurate the model can be
Training a Model
In Teachable Machine
- Create a new Image Project
- Rename the classes to the two scenarios you will have your model differentiate between
- Turn on webcam and click Hold to Record until you have a few hundred image samples in each test cases
→ Try to take the same photos but in different angles, positions, lighting, etc.
→ A lot of data is good but it's useless if they're all the exact same!
- Train the model