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.

Deployment methods for machine learning models

Deployment methods for machine learning models

From the course: Giving Computers Vision

Deployment methods for machine learning models

- Yeah. Is there more roles? I mean, management and all that, but I mean, is there any other functional, like, this is my day-to-day job, writing code, doing things. - Well, there is this whole ML Ops thing right. - Okay - Which is similar to what I mentioned before, but it also, it is a continuous, they would implement things like continuous, continuous integration. - Hmm - When you create a new model, it has to go through a large test suite - Yeah. - To make sure that there are no biases. It has to be better than the previous model. - Yep. - So all those things - can't break anything - Cannot break anything. - Right - So it goes through and, - Wow and it works on devices. - Yeah, cause you see cause that's massive. Cause now, now you're doing true and, and testing. - Yes. - Right. Yeah, yes it has to be better remove biases, blah, blah, blah. But, but it also. - From the onset it cannot screw up the…

Contents