From the course: Giving Computers Vision

Unlock this course with a free trial

Join today to access over 23,200 courses taught by industry experts.

How algorithms optimize themselves

How algorithms optimize themselves

From the course: Giving Computers Vision

Start my 1-month free trial Buy for my team

How algorithms optimize themselves

“

- [Interviewer] So, if you have to process a lot of stuff or train algorithms, GPUs are going to be your friend, and then but at the time of, let's say, using a model to determine something on a smaller set, you're going to probably save a lot of money just by going the CPU route? - [Interviewee] Yes. - [Interviewer] And do you tell, so if I'm writing in Python here and I'm saying, "Hey, go process this image" or let's say, a website is calling it and it's running its algorithm to determine something and then tell me I dunno, like, let's say, identifying an ID, the age of a person based on a picture they got, I'm going to use a CPU there, and now do I tell it that, or does it automatically there's maybe under the hood PyTorch or whatever thing I'm using says, "Oh, I'm just going to go CPU because it's quick and easy." And how do you figure that out? - [Interviewee] So deployment is a whole another ballgame, but let me…

Contents