def main():
srnet = nn.SRNET()
dataset = dman.DataSet()
sess = tf.compat.v1.InteractiveSession()
sess.run(tf.compat.v1.global_variables_initializer())
saver = tf.compat.v1.train.Saver()
# tfp.training(sess=sess, neuralnet=srnet, saver=saver, dataset=dataset, epochs=FLAGS.epoch, batch_size=FLAGS.batch)
tfp.validation(sess=sess, neuralnet=srnet, saver=saver, dataset=dataset)
def main():
dataset = dman.DataSet(setname=FLAGS.setname, tr_ratio=FLAGS.tr_ratio)
neuralnet = nn.ConvNet(data_dim=dataset.data_dim, channel=dataset.channel, num_class=dataset.num_class, learning_rate=FLAGS.lr)
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
tfp.training(sess=sess, neuralnet=neuralnet, saver=saver, dataset=dataset, epochs=FLAGS.epoch, batch_size=FLAGS.batch, dropout=FLAGS.dropout)
tfp.validation(sess=sess, neuralnet=neuralnet, saver=saver, dataset=dataset)
def main():
if (not (torch.cuda.is_available())): FLAGS.ngpu = 0
device = torch.device("cuda" if (
torch.cuda.is_available() and FLAGS.ngpu > 0) else "cpu")
srnet = nn.NeuralNet(device=device, ngpu=FLAGS.ngpu)
dataset = dman.DataSet()
solver.training(neuralnet=srnet,
dataset=dataset,
epochs=FLAGS.epoch,
batch_size=FLAGS.batch)
solver.validation(neuralnet=srnet, dataset=dataset)
def main():
srnet = nn.SRNET()
dataset = dman.DataSet()
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
tfp.training(sess=sess,
neuralnet=srnet,
saver=saver,
dataset=dataset,
iteration=int(FLAGS.iter),
batch_size=FLAGS.batch)
tfp.validation(sess=sess, neuralnet=srnet, saver=saver, dataset=dataset)
def main():
training_keys = ['100']
dataset = dman.DataSet(key_tr=training_keys)
lstm = nn.LSTM_Model(batch_size=FLAGS.batch, data_dim=dataset.data_dim)
sess = tf.InteractiveSession()
sess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
tfp.training(sess=sess,
neuralnet=lstm,
saver=saver,
dataset=dataset,
batch_size=FLAGS.batch,
sequence_length=FLAGS.trainlen,
iteration=FLAGS.iter)
tfp.validation(sess=sess,
neuralnet=lstm,
saver=saver,
dataset=dataset,
sequence_length=FLAGS.testlen)