From the course: Leveraging Cloud-Based Machine Learning on Google Cloud Platform: Real World Applications

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Vision AI training

Vision AI training

- [Narrator] So training the models in Vision is fairly straight forward. We just need to provide the images, and those images need to be labeled as to what they are. Then we're able to enter those images directly into a training routine. So it looks at the images, looks at the patterns in the images, and is able to discern what the images are and create a knowledge base on the differences that we're showing in what those images are representing. Now for example, a fraudulent check versus a non-fraudulent check, a type of cloud, a type of plant, all these things are possible within Vision. We're able to leverage the tests and evaluate function that you'll see in the demonstration which allows us to test the model and evaluate what's coming out in terms of predictability. In other words, we can figure out whether the model is working well or not working well. Then most importantly, predictions, the ability to enter in…

Contents