def evaluate():
with tf.Graph().as_default():
# Get images and labels for aurora.
images, labels = dnn.inputs(True)
# Build a Graph that computes the logits predictions from the
# inference model.
logits = dnn.inference(images)
# Calculate predictions.
top_k_op = tf.nn.in_top_k(logits, labels, 1)
# Restore the moving average version of the learned variables for eval.
variable_averages = tf.train.ExponentialMovingAverage(
dnn.MOVING_AVERAGE_DECAY)
# variables_to_restore = variable_averages.variables_to_restore()
# saver = tf.train.Saver(variables_to_restore)
saver = tf.train.Saver()
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
graph_def = tf.get_default_graph().as_graph_def()
summary_writer = tf.train.SummaryWriter(FLAGS.eval_dir,
graph_def=graph_def)
while True:
eval_once(saver, summary_writer, top_k_op, summary_op)
if FLAGS.run_once:
break
time.sleep(FLAGS.eval_interval_secs)
def train():
"""Train aurora images for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Get images and labels for aurora.
images, labels = dnn1.distorted_inputs()
# Build a Graph that computes the logits predictions from the
# inference model.
logits = dnn1.inference(images)
# Calculate loss.
loss = dnn1.loss(logits, labels)
# Build a Graph that trains the model with one batch of examples and
# updates the model parameters.
train_op = dnn1.train(loss, global_step)
# Create a saver.
saver = tf.train.Saver(tf.all_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
# aggregation_method = 2
### start session
config = tf.ConfigProto()
# config.gpu_options.per_process_gpu_memory_fraction=0.98
config.gpu_options.allocator_type = "BFC"
config.log_device_placement = FLAGS.log_device_placement
# sess = tf.Session(config=tf.ConfigProto(log_device_placement=FLAGS.log_device_placement))
sess = tf.Session(config=config)
sess.run(init)
# Start the queue runners.
tf.train.start_queue_runners(sess=sess)
summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
graph_def=sess.graph_def)
for step in xrange(FLAGS.max_steps):
start_time = time.time()
_, loss_value = sess.run([train_op, loss])
duration = time.time() - start_time
assert not np.isnan(loss_value), 'Model diverged with loss = NaN'
if step % 10 == 0:
num_examples_per_step = FLAGS.batch_size
examples_per_sec = num_examples_per_step / duration
sec_per_batch = float(duration)
format_str = (
'%s: step %d, loss = %.2f (%.1f examples/sec; %.3f '
'sec/batch)')
print(format_str % (datetime.now(), step, loss_value,
examples_per_sec, sec_per_batch))
if step % 100 == 0:
summary_str = sess.run(summary_op)
summary_writer.add_summary(summary_str, step)
# Save the model checkpoint periodically.
if step % 1000 == 0 or (step + 1) == FLAGS.max_steps:
checkpoint_path = os.path.join(FLAGS.train_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
def train():
"""Train aurora images for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Get images and labels for aurora.
images, labels = dnn1.distorted_inputs()
# Build a Graph that computes the logits predictions from the
# inference model.
logits = dnn1.inference(images)
# Calculate loss.
loss = dnn1.loss(logits, labels)
# Build a Graph that trains the model with one batch of examples and
# updates the model parameters.
train_op = dnn1.train(loss, global_step)
# Create a saver.
saver = tf.train.Saver(tf.all_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
# Build an initialization operation to run below.
init = tf.initialize_all_variables()
# Start running operations on the Graph.
# aggregation_method = 2
### start session
config = tf.ConfigProto()
# config.gpu_options.per_process_gpu_memory_fraction=0.98
config.gpu_options.allocator_type = "BFC"
config.log_device_placement = FLAGS.log_device_placement
# sess = tf.Session(config=tf.ConfigProto(log_device_placement=FLAGS.log_device_placement))
sess = tf.Session(config=config)
sess.run(init)
# Start the queue runners.
tf.train.start_queue_runners(sess=sess)
summary_writer = tf.train.SummaryWriter(FLAGS.train_dir,
graph_def=sess.graph_def)
for step in xrange(FLAGS.max_steps):
start_time = time.time()
_, loss_value = sess.run([train_op, loss])
duration = time.time() - start_time
assert not np.isnan(loss_value), 'Model diverged with loss = NaN'
if step % 10 == 0:
num_examples_per_step = FLAGS.batch_size
examples_per_sec = num_examples_per_step / duration
sec_per_batch = float(duration)
format_str = ('%s: step %d, loss = %.2f (%.1f examples/sec; %.3f '
'sec/batch)')
print (format_str % (datetime.now(), step, loss_value,
examples_per_sec, sec_per_batch))
if step % 100 == 0:
summary_str = sess.run(summary_op)
summary_writer.add_summary(summary_str, step)
# Save the model checkpoint periodically.
if step % 1000 == 0 or (step + 1) == FLAGS.max_steps:
checkpoint_path = os.path.join(FLAGS.train_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)
