def do_eval(sess):
"""Runs one evaluation against the full epoch of data.
Args:
sess: The session in which the model has been trained.
eval_correct: The Tensor that returns the number of correct predictions.
"""
left_batch_eval, right_batch_eval, lidar_batch_eval = simladar.inputs(
eval_data=True)
logits_eval = simladar.inference(left_batch_eval, right_batch_eval)
loss_eval = simladar.eval_loss(logits_eval, lidar_batch_eval)
# And run one epoch of eval.
rmse = sess.run(loss_eval)
return rmse
def evaluate():
"""Eval CIFAR-10 for a number of steps."""
# with tf.Graph().as_default(),tf.device('/cpu:0') as g:
with tf.device('/cpu:0') as g:
global_step = tf.Variable(0, trainable=False)
# # Get images and labels for CIFAR-10.
# eval_data = FLAGS.eval_data == 'test'
# images, labels = cifar10.inputs(eval_data=eval_data)
left_batch_validate, right_batch_validate, lidar_batch_validate = simladar.inputs(
eval_data=True)
# Build a Graph that computes the logits predictions from the
# inference model.
# logits = cifar10.inference(images)
keep_prob = tf.constant(1.) #dropout (keep probability)
pred_validate = simladar.inference(left_batch_validate,
right_batch_validate, keep_prob)
# Calculate predictions.
# top_k_op = tf.nn.in_top_k(logits, labels, 1)
valid_op = tf.sqrt(
tf.reduce_mean(
tf.square(tf.sub(lidar_batch_validate, pred_validate))))
# Restore the moving average version of the learned variables for eval.
variable_averages = tf.train.ExponentialMovingAverage(
simladar.MOVING_AVERAGE_DECAY)
variables_to_restore = variable_averages.variables_to_restore()
saver = tf.train.Saver(variables_to_restore)
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.merge_all_summaries()
summary_writer = tf.train.SummaryWriter(FLAGS.eval_dir, g)
while True:
eval_once(saver, summary_writer, valid_op, summary_op)
if FLAGS.run_once:
break
time.sleep(FLAGS.eval_interval_secs)
def train():
"""Train CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Get images and labels .
left_batch, right_batch, lidar_batch = simladar.inputs()
# Build a Graph that computes the logits predictions from the
# inference model.
logits = simladar.inference(left_batch, right_batch)
# Calculate loss.
loss = simladar.loss(logits, lidar_batch)
# Build a Graph that trains the model with one batch of examples and
# updates the model parameters.
train_op = simladar.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.
sess = tf.Session(config=tf.ConfigProto(
log_device_placement=FLAGS.log_device_placement))
sess.run(init)
# Start the queue runners.
tf.train.start_queue_runners(sess=sess)
summary_writer = tf.train.SummaryWriter(FLAGS.train_dir, sess.graph)
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:
# eval_rmse = do_eval(sess)
# print('eval rmse: ',eval_rmse )
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 CIFAR-10 for a number of steps."""
with tf.Graph().as_default():
global_step = tf.Variable(0, trainable=False)
# Get images and labels .
left_batch, right_batch, lidar_batch = simladar.inputs()
keep_prob = tf.constant(DROPOUT_PROB) #dropout (keep probability)
# Build a Graph that computes the logits predictions from the
# inference model.
logits = simladar.inference(left_batch, right_batch, keep_prob)
# Calculate loss.
loss = simladar.loss(logits, lidar_batch)
# Build a Graph that trains the model with one batch of examples and
# updates the model parameters.
train_op = simladar.train(loss, global_step)
# Create a saver.
saver = tf.train.Saver(tf.global_variables())
# Build the summary operation based on the TF collection of Summaries.
summary_op = tf.summary.merge_all()
# Build an initialization operation to run below.
init = tf.global_variables_initializer()
# Start running operations on the Graph.
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
config.log_device_placement = FLAGS.log_device_placement
sess = tf.Session(config=config)
sess.run(init)
# load model
checkpoint = tf.train.get_checkpoint_state(FLAGS.save_dir)
if checkpoint and checkpoint.model_checkpoint_path:
saver.restore(sess, checkpoint.model_checkpoint_path)
print("Successfully loaded:", checkpoint.model_checkpoint_path)
# print("global step: ", global_step.eval())
else:
print("Could not find old network weights")
# Start the queue runners.
tf.train.start_queue_runners(sess=sess)
# summary_writer = tf.train.SummaryWriter(FLAGS.save_dir, sess.graph)
summary_writer = tf.summary.FileWriter(FLAGS.save_dir, sess.graph)
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:
# eval_rmse = do_eval(sess)
# print('eval rmse: ',eval_rmse )
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.save_dir, 'model.ckpt')
saver.save(sess, checkpoint_path, global_step=step)