RaphaelMeudec/deblur-gan

Can I train the model with multi GPUs?

wiwengweng opened this issue · 2 comments

Hi, I have two Nvidia cards, but when training the model, I see only one card is in use, although both cards' GPU memory is occupied.

root@bogon:~/DeblurGAN# nvidia-smi 
Wed Aug  8 13:59:21 2018       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 390.67                 Driver Version: 390.67                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 108...  Off  | 00000000:02:00.0 Off |                  N/A |
| 53%   58C    P2   222W / 250W |  10793MiB / 11178MiB |     85%      Default |
+-------------------------------+----------------------+----------------------+
|   1  GeForce GTX 108...  Off  | 00000000:82:00.0 Off |                  N/A |
| 47%   25C    P8    10W / 250W |  10631MiB / 11178MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

You can use Keras's multi-gpu model.

@ravikt Thanks!