多GPU的数据训练，feed

 

images, labels = cifar10.distorted_inputs()
split_images = tf.split(images, FLAGS.num_gpus, 0)
split_labels = tf.split(labels, FLAGS.num_gpus, 0)

 

for i in xrange(FLAGS.num_gpus):
  with tf.device('/gpu:%d' % i):
    with tf.name_scope('%s_%d' % (cifar10.TOWER_NAME, i)) as scope:

      loss = tower_loss(scope, split_images[i], split_labels[i])

 

https://*.com/questions/34273951/tensorflow-multi-gpu-single-input-queue?

--