Memory usage in keras
I'm using Keras 2.0 and can't use the GPU when my model is in the training process. It gives me out of memory. How can I use GPU during the training process?
I'm using Keras 2.0 and can't use the GPU when my model is in the training process. It gives me out of memory. How can I use GPU during the training process?
If you use Tensorflow as backend the you can change the configuration.
import tensorflow as tf
from keras.backend.tensorflow_backend import set_session
config = tf.ConfigProto()
config.gpu_options.per_process_gpu_memory_fraction = 0.3
set_session(tf.Session(config=config))