hello,
I'm wondering how to create a machine learning program that recognizes people based on a name they type in, and remember that name even after the stop button is clicked
i suspect an easier way to do it would be by using the database extension, but i am just wondering how you would go about doing that?
thanks for your time 
You wouldn’t be able to use cookies or databases, it would have to be cloud vars.
cool 
how do you make a variable a cloud one?
i googled it and couldn't find anything
This isn't really a machine learning problem. Long before machine learning, people were using usernames to log into computers. It's just a question of maintaining a list of names.
Data in your project don't disappear because you click the stop button. That part's easy. What those other replies are about is how to handle the situation in which you want to remember the name after you quit Snap! and come back the next day.
If you just want to remember the name in the same browser on the same computer, you can use the database library, which uses browser localstore -- more or less cookies, but with slightly different rules. If you want to be able to use the same name when you're on a different computer, then you need some sort of cloud storage. That's not something we offer, but various users have established semi-clandestine cloud variable servers. Search the forum for "cloud variables."
If you want to store data on the web and have it be recognized from another computer. but i think using the database library would be better for this since cloud vars a) need their own web server to work, b) are really slow for getting data, and c) aren't really needed for this purpose.
I made a simple AI a few months ago. You can check it out here: How to make an enemy AI agent for your top-down games
They mean machine learning, not just an algorithm.
Oh. Well, I DID make an AI Guessing Game using Python.
Source Code: AI Guessing Game (SOURCE CODE) - Google Docs
You have to find one on Snap!, find a free cloud server, or buy one and make the blocks.
Not only is this pointless shilling of a non-Snap! project, it's still not related to the topic, which is just about storing data in variables or the database library. (Also my bad; I should've read the topic again before making assumptions.)
25 GB would never be enough, for machine learning to guess an image correctly 60% of the time, you would need 10,000 images of each item, 25 GB is like IDK, 50 to 100 images?
You guys should look up the shared projects and the Snap!Con talks by Ken Kahn, @toontalk, who has been working in this area since forever and has projects both self-contained in your computer and using Google cloud services.
How do I get the recordings of Snap!Con 2021?
@slate_technologies @helicoptur : Instead of flagging each other's posts, how about you talk nicely to each other and resolve your differences?
I never flagged his posts.
No, I guess you're right, you just posted
... which is more or less the same as flagging it. But I don't want to start a new fight, I want to eliminate the existing one.