Open ivyteng opened 6 years ago
Easiest way is to specify it command line when starting the process via
CUDA_VISIBLE_DEVICES=2,7 python predict.py
GPU ids are 0 based.
Otherwise define the model under a block
with tf.device('/device:GPU:2'):
model = VGG16()```
by using Tensor Flow graphs you can specify a single GPU when there are more than one GPU.