amuamushu / adv_avod_ssn

Deep Sensor Fusion for Single Source Robustness
https://ayushmore.github.io/2022-03-07-improving-robustness-via-adversarial-training/
0 stars 0 forks source link

Check that GPU is running for tensorflow. #10

Closed amuamushu closed 2 years ago

amuamushu commented 2 years ago

Checking that GPU is actually running can be made using nvdia smi.

Our mentor Lily suggested running the code and opening another terminal to see if GPU enabler is really working.

amuamushu commented 2 years ago

It seems the GPU isn't really being used.

Running nvidia smi before running run.py: image

Running nvidia smi during the run of run.py: image

Though, the GPU device was mentioned during the run 2022-02-01 15:38:26.030182: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1510] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 9659 MB memory: -> device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:b2:00.0, compute capability: 7.5

amuamushu commented 2 years ago

Talked to a TA about this and he mentioned the bottom table being blank is just a bug with nvidia-smi and that the GPU is actually being used.

Talked to another TA about optimizing existing code. (documented in https://github.com/amuamushu/adv_avod_ssn/issues/6#issuecomment-1029469546)