The computer restarts large mini-parties in TensorFlow

I am running TensorFlow for Windows with a Titan X GPU (12 GB of memory). When I try to organize a network for 256X256X1 images with mini-packets larger than 50 images, my computer will just work and restart automatically. With smaller mini-parties, it works great. Any clues on what could be causing this?

+5
source share
1 answer

I saw similar issues discussed in some gaming forums where the PC just shut down when the GPU was under heavy load. The reason usually was that the GPU consumed more power than the power supply could work. Check for example. here or here . So is it worth investigating if your power supply is guilty.

Edit: Maybe SpeedFan can help you debug this - it can show both voltages and temperature sensors, which will also tell you if your computer is overheating (I never used the tool myself, and I'm also not connected to him, just found it online).

0
source

Source: https://habr.com/ru/post/1261550/


All Articles