DeepTrial / Retina-VesselNet

A Simple U-net model for Retinal Blood Vessel Segmentation based on tensorflow2
305 stars 76 forks source link

Server Specification #1

Closed hesoyamcode closed 6 years ago

hesoyamcode commented 6 years ago

Hi Xing ChenCong

Previously big thanks for the good code! I have questions what kind of server / compute you use to achieve the train and test result? And how long it takes?

Thanks

DeepTrial commented 6 years ago

Hi, I trained it on my own laptop(msi GL62m 7rd) with one GTX 1050 2g GPU.I trained for about 2 hours and the loss vanished