Playing with a DL build design (https://pcpartpicker.com/list/MK2wxY ), excited about NVM SSDs. Will allow devs to load more 100s of MBs of JS frameworks!
Is your data preprocessing pipeline single core bound? Is it easily parallelizable?
-
-
Not particularly relevant to RL/MCMC/ABC; and image CNNs don't require much preprocessing other than rescaling which can be done in advance.
-
eg I was working with GAN on v slow HDD, and bottlenecked on disk reads; so I rescaled down manually & used `find` to force JPGs into cache.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.