Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
-
At my old job (where I did ML):
Company: Okay we have a small server in the basement with 128 gb ram. You need a GPU for training, let’s get a 1080Ti.
Receives it after ordering
Company: oh shoot, the GPU is to big for the small server. Anyone at the office who wants a free 1080Ti?
My colleague got it (he had been there longer than me).
Later they talked about buying a new larger server with 2xV100. One of my colleagues there once told me, that at that specific company, money is just something that you spend... Ridiculous
Me: I need a Windows PC to test my code on.
Server dude: We don't have any free, but I can setup an old server...
Oh no GPU what have you done!
That's not AI.
Me, starting my internship in ML.. coworkers come to me asking what computer I need:
Me: Well, something more powerful than this i3, and most importantly some kind of GPU for training.
Them: Ok, what kind of GPU?
Me: Well, a 1080 or 2080 should be more than enough and good performance for the price.
Them: Oh.. We were more thinking about a Tesla V100 or something like that!
Me: (internally) WTF this costs more than what you'll pay me for the internship, this is so cool. (to them) Oh, yes, why not, great perfomance, blah blah blah.
I would prefer them to pay me more, but at least they're not going to hold me down with bad components! Nothing to rant about for now.. Hope it'll stay the same ^^
rant
ml
gpu
overkill
canikeepit?