Google Cloud AutoML

Today, i attended the Google Developers Group here in Cebu,Philippines and one of the speaker introduced this cool product.

AutoML is categorized into three parts. These are Vision, Natural Language, and Translations.

The AI (Artificial Intelligence) & Machine Learning will be train by you as the developer, these Machine Learning must enhance their skills in order to identify a certain thing.

I will give you an example of AutoML Vision:

Photo above, the Machine Learning tells the user that it is a human.

By how?

The developer must enhance the ML by uploading different kinds of picture of a human, more or less 200 to 1000 photo would be enough to train the skills of the ML.

And because of that, when the user put some photos of a human, the ML will tell a high percentage that it is a human that is because the skills was enhanced.