def main(_):
clone_batch_size = FLAGS.batch_size
steps_per_epoch = FLAGS.training_number_of_steps
m_epoch = FLAGS.epoch
train_dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.train_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=clone_batch_size,
crop_size=FLAGS.train_crop_size,
min_scale_factor=FLAGS.min_scale_factor,
max_scale_factor=FLAGS.max_scale_factor,
num_readers=2,
is_training=True,
should_shuffle=True,
should_repeat=True)
train_iterator = train_dataset.get_one_shot_iterator()
test_dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.test_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=clone_batch_size,
crop_size=FLAGS.train_crop_size,
min_scale_factor=FLAGS.min_scale_factor,
max_scale_factor=FLAGS.max_scale_factor,
num_readers=2,
is_training=False,
should_shuffle=True,
should_repeat=True)
test_iterator = test_dataset.get_one_shot_iterator()
# build model
my_model, init_e = build_model()
# train
train(my_model,
train_iterator,
steps_per_epoch,
m_epoch,
initial_epoch=init_e)
# evaluate and predict
eval_and_predict(my_model, test_iterator)
# summary and save
my_model.summary()
def main(_):
clone_batch_size = 8
# steps_per_epoch = int(1800 / clone_batch_size)
# m_epoch = int((FLAGS.training_number_of_steps+1800) / 1800)
steps_per_epoch = 600 * 4
m_epoch = 1
train_dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.train_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=clone_batch_size,
crop_size=FLAGS.train_crop_size,
min_scale_factor=FLAGS.min_scale_factor,
max_scale_factor=FLAGS.max_scale_factor,
num_readers=2,
is_training=True,
should_shuffle=False,
should_repeat=False)
train_iterator = train_dataset.get_one_shot_iterator()
my_train(m_epoch,
steps_per_epoch,
batch_handler=handle_batch,
iterator=train_iterator)
def testPascalVocSegTestData(self):
dataset = data_generator.Dataset(
dataset_name='pascal_voc_seg',
split_name='val',
dataset_dir=
'deeplab/testing/pascal_voc_seg',
batch_size=1,
crop_size=[3, 3], # Use small size for testing.
min_resize_value=3,
max_resize_value=3,
resize_factor=None,
min_scale_factor=0.01,
max_scale_factor=2.0,
scale_factor_step_size=0.25,
is_training=False,
model_variant='mobilenet_v2')
self.assertAllEqual(dataset.num_of_classes, 21)
self.assertAllEqual(dataset.ignore_label, 255)
num_of_images = 3
with self.test_session() as sess:
iterator = dataset.get_one_shot_iterator()
for i in range(num_of_images):
batch = iterator.get_next()
batch, = sess.run([batch])
image_attributes = _get_attributes_of_image(i)
self.assertEqual(batch[common.HEIGHT][0], image_attributes.height)
self.assertEqual(batch[common.WIDTH][0], image_attributes.width)
self.assertEqual(batch[common.IMAGE_NAME][0],
image_attributes.image_name.encode())
# All data have been read.
with self.assertRaisesRegexp(tf.errors.OutOfRangeError, ''):
sess.run([iterator.get_next()])
def main(unused_argv):
tf.logging.set_verbosity(tf.logging.INFO)
# Get dataset-dependent information.
dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.vis_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=FLAGS.vis_batch_size,
crop_size=[int(sz) for sz in FLAGS.vis_crop_size],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
resize_factor=FLAGS.resize_factor,
model_variant=FLAGS.model_variant,
is_training=False,
should_shuffle=False,
should_repeat=False)
train_id_to_eval_id = None
if dataset.dataset_name == data_generator.get_cityscapes_dataset_name():
tf.logging.info('Cityscapes requires converting train_id to eval_id.')
train_id_to_eval_id = _CITYSCAPES_TRAIN_ID_TO_EVAL_ID
# Prepare for visualization.
tf.gfile.MakeDirs(FLAGS.vis_logdir)
save_dir = os.path.join(FLAGS.vis_logdir, _SEMANTIC_PREDICTION_SAVE_FOLDER)
tf.gfile.MakeDirs(save_dir)
raw_save_dir = os.path.join(
FLAGS.vis_logdir, _RAW_SEMANTIC_PREDICTION_SAVE_FOLDER)
tf.gfile.MakeDirs(raw_save_dir)
tf.logging.info('Visualizing on %s set', FLAGS.vis_split)
with tf.Graph().as_default():
samples = dataset.get_one_shot_iterator().get_next()
model_options = common.ModelOptions(
outputs_to_num_classes={common.OUTPUT_TYPE: dataset.num_of_classes},
crop_size=[int(sz) for sz in FLAGS.vis_crop_size],
atrous_rates=FLAGS.atrous_rates,
output_stride=FLAGS.output_stride)
if tuple(FLAGS.eval_scales) == (1.0,):
tf.logging.info('Performing single-scale test.')
predictions = model.predict_labels(
samples[common.IMAGE],
model_options=model_options,
image_pyramid=FLAGS.image_pyramid)
else:
tf.logging.info('Performing multi-scale test.')
if FLAGS.quantize_delay_step >= 0:
raise ValueError(
'Quantize mode is not supported with multi-scale test.')
predictions = model.predict_labels_multi_scale(
samples[common.IMAGE],
model_options=model_options,
eval_scales=FLAGS.eval_scales,
add_flipped_images=FLAGS.add_flipped_images)
predictions = predictions[common.OUTPUT_TYPE]
if FLAGS.min_resize_value and FLAGS.max_resize_value:
# Only support batch_size = 1, since we assume the dimensions of original
# image after tf.squeeze is [height, width, 3].
assert FLAGS.vis_batch_size == 1
# Reverse the resizing and padding operations performed in preprocessing.
# First, we slice the valid regions (i.e., remove padded region) and then
# we resize the predictions back.
original_image = tf.squeeze(samples[common.ORIGINAL_IMAGE])
original_image_shape = tf.shape(original_image)
predictions = tf.slice(
predictions,
[0, 0, 0],
[1, original_image_shape[0], original_image_shape[1]])
resized_shape = tf.to_int32([tf.squeeze(samples[common.HEIGHT]),
tf.squeeze(samples[common.WIDTH])])
predictions = tf.squeeze(
tf.image.resize_images(tf.expand_dims(predictions, 3),
resized_shape,
method=tf.image.ResizeMethod.NEAREST_NEIGHBOR,
align_corners=True), 3)
tf.train.get_or_create_global_step()
if FLAGS.quantize_delay_step >= 0:
contrib_quantize.create_eval_graph()
num_iteration = 0
max_num_iteration = FLAGS.max_number_of_iterations
checkpoints_iterator = contrib_training.checkpoints_iterator(
FLAGS.checkpoint_dir, min_interval_secs=FLAGS.eval_interval_secs)
for checkpoint_path in checkpoints_iterator:
num_iteration += 1
tf.logging.info(
'Starting visualization at ' + time.strftime('%Y-%m-%d-%H:%M:%S',
time.gmtime()))
tf.logging.info('Visualizing with model %s', checkpoint_path)
scaffold = tf.train.Scaffold(init_op=tf.global_variables_initializer())
session_creator = tf.train.ChiefSessionCreator(
scaffold=scaffold,
master=FLAGS.master,
checkpoint_filename_with_path=checkpoint_path)
with tf.train.MonitoredSession(
session_creator=session_creator, hooks=None) as sess:
batch = 0
image_id_offset = 0
while not sess.should_stop():
tf.logging.info('Visualizing batch %d', batch + 1)
_process_batch(sess=sess,
original_images=samples[common.ORIGINAL_IMAGE],
semantic_predictions=predictions,
image_names=samples[common.IMAGE_NAME],
image_heights=samples[common.HEIGHT],
image_widths=samples[common.WIDTH],
image_id_offset=image_id_offset,
save_dir=save_dir,
raw_save_dir=raw_save_dir,
train_id_to_eval_id=train_id_to_eval_id)
image_id_offset += FLAGS.vis_batch_size
batch += 1
tf.logging.info(
'Finished visualization at ' + time.strftime('%Y-%m-%d-%H:%M:%S',
time.gmtime()))
if max_num_iteration > 0 and num_iteration >= max_num_iteration:
break
def main(unused_argv):
tf.logging.set_verbosity(tf.logging.INFO)
dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.eval_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=FLAGS.eval_batch_size,
crop_size=[int(sz) for sz in FLAGS.eval_crop_size],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
resize_factor=FLAGS.resize_factor,
model_variant=FLAGS.model_variant,
num_readers=2,
is_training=False,
should_shuffle=False,
should_repeat=False)
tf.gfile.MakeDirs(FLAGS.eval_logdir)
tf.logging.info('Evaluating on %s set', FLAGS.eval_split)
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=1.0)
session_config = tf.ConfigProto(allow_soft_placement=True,
log_device_placement=False,
gpu_options=gpu_options)
#session_config.gpu_options.allow_growth = True
with tf.Graph().as_default():
samples = dataset.get_one_shot_iterator().get_next()
#print(samples[common.IMAGE_NAME])
model_options = common.ModelOptions(
outputs_to_num_classes={
common.OUTPUT_TYPE: dataset.num_of_classes
},
crop_size=[int(sz) for sz in FLAGS.eval_crop_size],
atrous_rates=FLAGS.atrous_rates,
output_stride=FLAGS.output_stride)
# Set shape in order for tf.contrib.tfprof.model_analyzer to work properly.
samples[common.IMAGE].set_shape([
FLAGS.eval_batch_size,
int(FLAGS.eval_crop_size[0]),
int(FLAGS.eval_crop_size[1]), 3
])
if tuple(FLAGS.eval_scales) == (1.0, ):
tf.logging.info('Performing single-scale test.')
predictions, logits = model.predict_labels(
samples[common.IMAGE],
model_options,
image_pyramid=FLAGS.image_pyramid,
skips=FLAGS.skips)
else:
tf.logging.info('Performing multi-scale test.')
if FLAGS.quantize_delay_step >= 0:
raise ValueError(
'Quantize mode is not supported with multi-scale test.')
predictions = model.predict_labels_multi_scale(
samples[common.IMAGE],
model_options=model_options,
skips=FLAGS.skips,
eval_scales=FLAGS.eval_scales,
add_flipped_images=FLAGS.add_flipped_images)
predictions = predictions[common.OUTPUT_TYPE]
predictions = tf.reshape(predictions, shape=[-1])
labels = tf.reshape(samples[common.LABEL], shape=[-1])
weights = tf.to_float(tf.not_equal(labels, dataset.ignore_label))
# Set ignore_label regions to label 0, because metrics.mean_iou requires
# range of labels = [0, dataset.num_classes). Note the ignore_label regions
# are not evaluated since the corresponding regions contain weights = 0.
labels = tf.where(tf.equal(labels, dataset.ignore_label),
tf.zeros_like(labels), labels)
predictions_tag = 'miou'
for eval_scale in FLAGS.eval_scales:
predictions_tag += '_' + str(eval_scale)
if FLAGS.add_flipped_images:
predictions_tag += '_flipped'
# Define the evaluation metric.
metric_map = {}
# to remove "predictions out of bound error"
indices = tf.squeeze(
tf.where(tf.less_equal(labels, dataset.num_of_classes - 1)), 1)
labels_ind = tf.cast(tf.gather(labels, indices), tf.int32)
predictions_ind = tf.gather(predictions, indices)
# end of insert
miou, update_miou = tf.metrics.mean_iou(labels_ind,
predictions_ind,
dataset.num_of_classes,
weights=weights,
name="mean_iou")
tf.summary.scalar(predictions_tag, miou)
# Define the evaluation metric IOU for individual classes
iou_v, update_op = my_metrics.iou(labels_ind,
predictions_ind,
dataset.num_of_classes,
weights=weights)
for index in range(0, dataset.num_of_classes):
metric_map['class_' + str(index) + '_iou'] = (iou_v[index],
update_op[index])
tf.summary.scalar('class_' + str(index) + '_iou', iou_v[index])
# Confusion matrix save hook. It updates the confusion matrix on tensorboard at the end of eval loop.
confusionMatrixSaveHook = confusion_matrix.SaverHook(
labels=['BG', 'water', 'ice', 'snow', 'clutter'],
confusion_matrix_tensor_name='mean_iou/total_confusion_matrix',
summary_writer=tf.summary.FileWriterCache.get(
str(FLAGS.eval_logdir)))
summary_op = tf.summary.merge_all()
summary_hook = tf.contrib.training.SummaryAtEndHook(
log_dir=FLAGS.eval_logdir, summary_op=summary_op)
hooks = [summary_hook, confusionMatrixSaveHook]
num_eval_iters = None
if FLAGS.max_number_of_evaluations > 0:
num_eval_iters = FLAGS.max_number_of_evaluations
if FLAGS.quantize_delay_step >= 0:
tf.contrib.quantize.create_eval_graph()
tf.contrib.training.evaluate_repeatedly(
master=FLAGS.master,
checkpoint_dir=FLAGS.checkpoint_dir,
eval_ops=[update_miou, update_op],
max_number_of_evaluations=num_eval_iters,
hooks=hooks,
eval_interval_secs=FLAGS.eval_interval_secs)
def main(unused_argv):
tf.logging.set_verbosity(tf.logging.INFO)
tf.gfile.MakeDirs(FLAGS.train_logdir)
tf.logging.info('Training on %s set', FLAGS.train_split)
graph = tf.Graph()
with graph.as_default():
with tf.device(
tf.train.replica_device_setter(ps_tasks=FLAGS.num_ps_tasks)):
assert FLAGS.train_batch_size % FLAGS.num_clones == 0, (
'Training batch size not divisble by number of clones (GPUs).')
clone_batch_size = FLAGS.train_batch_size // FLAGS.num_clones
dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.train_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=clone_batch_size,
crop_size=[int(sz) for sz in FLAGS.train_crop_size],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
resize_factor=FLAGS.resize_factor,
min_scale_factor=FLAGS.min_scale_factor,
max_scale_factor=FLAGS.max_scale_factor,
scale_factor_step_size=FLAGS.scale_factor_step_size,
model_variant=FLAGS.model_variant,
num_readers=2,
is_training=True,
should_shuffle=True,
should_repeat=True)
train_tensor, summary_op = _train_deeplab_model(
dataset.get_one_shot_iterator(), dataset.num_of_classes,
dataset.ignore_label)
# Soft placement allows placing on CPU ops without GPU implementation.
session_config = tf.ConfigProto(allow_soft_placement=True,
log_device_placement=False)
last_layers = model.get_extra_layer_scopes(
FLAGS.last_layers_contain_logits_only)
init_fn = None
if FLAGS.tf_initial_checkpoint:
init_fn = train_utils.get_model_init_fn(
FLAGS.train_logdir,
FLAGS.tf_initial_checkpoint,
FLAGS.initialize_last_layer,
last_layers,
ignore_missing_vars=True)
scaffold = tf.train.Scaffold(
init_fn=init_fn,
summary_op=summary_op,
)
stop_hook = tf.train.StopAtStepHook(
last_step=FLAGS.training_number_of_steps)
profile_dir = FLAGS.profile_logdir
if profile_dir is not None:
tf.gfile.MakeDirs(profile_dir)
with tf.contrib.tfprof.ProfileContext(enabled=profile_dir
is not None,
profile_dir=profile_dir):
with tf.train.MonitoredTrainingSession(
master=FLAGS.master,
is_chief=(FLAGS.task == 0),
config=session_config,
scaffold=scaffold,
checkpoint_dir=FLAGS.train_logdir,
summary_dir=FLAGS.train_logdir,
log_step_count_steps=FLAGS.log_steps,
save_summaries_steps=FLAGS.save_summaries_secs,
save_checkpoint_secs=FLAGS.save_interval_secs,
hooks=[stop_hook]) as sess:
while not sess.should_stop():
sess.run([train_tensor])
def main(unused_argv):
tf.logging.set_verbosity(tf.logging.INFO)
dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.eval_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=FLAGS.eval_batch_size,
crop_size=[int(sz) for sz in FLAGS.eval_crop_size],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
resize_factor=FLAGS.resize_factor,
model_variant=FLAGS.model_variant,
num_readers=2,
is_training=False,
should_shuffle=False,
should_repeat=False)
tf.gfile.MakeDirs(FLAGS.eval_logdir)
tf.logging.info('Evaluating on %s set', FLAGS.eval_split)
with tf.Graph().as_default():
samples = dataset.get_one_shot_iterator().get_next()
model_options = common.ModelOptions(
outputs_to_num_classes={
common.OUTPUT_TYPE: dataset.num_of_classes
},
crop_size=[int(sz) for sz in FLAGS.eval_crop_size],
atrous_rates=FLAGS.atrous_rates,
output_stride=FLAGS.output_stride)
# Set shape in order for tf.contrib.tfprof.model_analyzer to work properly.
samples[common.IMAGE].set_shape([
FLAGS.eval_batch_size,
int(FLAGS.eval_crop_size[0]),
int(FLAGS.eval_crop_size[1]), 3
])
if tuple(FLAGS.eval_scales) == (1.0, ):
tf.logging.info('Performing single-scale test.')
predictions = model_func.predict_labels(
samples[common.IMAGE],
model_options,
image_pyramid=FLAGS.image_pyramid)
else:
tf.logging.info('Performing multi-scale test.')
if FLAGS.quantize_delay_step >= 0:
raise ValueError(
'Quantize mode is not supported with multi-scale test.')
predictions = model_func.predict_labels_multi_scale(
samples[common.IMAGE],
model_options=model_options,
eval_scales=FLAGS.eval_scales,
add_flipped_images=FLAGS.add_flipped_images)
predictions = predictions[common.OUTPUT_TYPE]
predictions = tf.reshape(predictions, shape=[-1])
labels = tf.reshape(samples[common.LABEL], shape=[-1])
weights = tf.to_float(tf.not_equal(labels, dataset.ignore_label))
# Set ignore_label regions to label 0, because metrics.mean_iou requires
# range of labels = [0, dataset.num_classes). Note the ignore_label regions
# are not evaluated since the corresponding regions contain weights = 0.
labels = tf.where(tf.equal(labels, dataset.ignore_label),
tf.zeros_like(labels), labels)
predictions_tag = 'miou'
for eval_scale in FLAGS.eval_scales:
predictions_tag += '_' + str(eval_scale)
if FLAGS.add_flipped_images:
predictions_tag += '_flipped'
# Define the evaluation metric.
metric_map = {}
num_classes = dataset.num_of_classes
metric_map['eval/%s_overall' % predictions_tag] = tf.metrics.mean_iou(
labels=labels,
predictions=predictions,
num_classes=num_classes,
weights=weights)
# IoU for each class.
one_hot_predictions = tf.one_hot(predictions, num_classes)
one_hot_predictions = tf.reshape(one_hot_predictions,
[-1, num_classes])
one_hot_labels = tf.one_hot(labels, num_classes)
one_hot_labels = tf.reshape(one_hot_labels, [-1, num_classes])
for c in range(num_classes):
predictions_tag_c = '%s_class_%d' % (predictions_tag, c)
tp, tp_op = tf.metrics.true_positives(
labels=one_hot_labels[:, c],
predictions=one_hot_predictions[:, c],
weights=weights)
fp, fp_op = tf.metrics.false_positives(
labels=one_hot_labels[:, c],
predictions=one_hot_predictions[:, c],
weights=weights)
fn, fn_op = tf.metrics.false_negatives(
labels=one_hot_labels[:, c],
predictions=one_hot_predictions[:, c],
weights=weights)
tp_fp_fn_op = tf.group(tp_op, fp_op, fn_op)
iou = tf.where(tf.greater(tp + fn, 0.0), tp / (tp + fn + fp),
tf.constant(np.NaN))
metric_map['eval/%s' % predictions_tag_c] = (iou, tp_fp_fn_op)
(metrics_to_values,
metrics_to_updates) = contrib_metrics.aggregate_metric_map(metric_map)
summary_ops = []
for metric_name, metric_value in six.iteritems(metrics_to_values):
op = tf.summary.scalar(metric_name, metric_value)
op = tf.Print(op, [metric_value], metric_name)
summary_ops.append(op)
summary_op = tf.summary.merge(summary_ops)
summary_hook = contrib_training.SummaryAtEndHook(
log_dir=FLAGS.eval_logdir, summary_op=summary_op)
hooks = [summary_hook]
num_eval_iters = None
if FLAGS.max_number_of_evaluations > 0:
num_eval_iters = FLAGS.max_number_of_evaluations
if FLAGS.quantize_delay_step >= 0:
contrib_quantize.create_eval_graph()
contrib_tfprof.model_analyzer.print_model_analysis(
tf.get_default_graph(),
tfprof_options=contrib_tfprof.model_analyzer.
TRAINABLE_VARS_PARAMS_STAT_OPTIONS)
contrib_tfprof.model_analyzer.print_model_analysis(
tf.get_default_graph(),
tfprof_options=contrib_tfprof.model_analyzer.FLOAT_OPS_OPTIONS)
contrib_training.evaluate_repeatedly(
checkpoint_dir=FLAGS.checkpoint_dir,
master=FLAGS.master,
eval_ops=list(metrics_to_updates.values()),
max_number_of_evaluations=num_eval_iters,
hooks=hooks,
eval_interval_secs=FLAGS.eval_interval_secs)
def main(unused_argv):
tf.logging.set_verbosity(tf.logging.INFO)
# Set up deployment (i.e., multi-GPUs and/or multi-replicas).
config = model_deploy.DeploymentConfig(num_clones=FLAGS.num_clones,
clone_on_cpu=FLAGS.clone_on_cpu,
replica_id=FLAGS.task,
num_replicas=FLAGS.num_replicas,
num_ps_tasks=FLAGS.num_ps_tasks)
# Split the batch across GPUs.
assert FLAGS.train_batch_size % config.num_clones == 0, (
'Training batch size not divisble by number of clones (GPUs).')
clone_batch_size = FLAGS.train_batch_size // config.num_clones
tf.gfile.MakeDirs(FLAGS.train_logdir)
tf.logging.info('Training on %s set', FLAGS.train_split)
with tf.Graph().as_default() as graph:
with tf.device(config.inputs_device()):
dataset = data_generator.Dataset(
dataset_name=FLAGS.dataset,
split_name=FLAGS.train_split,
dataset_dir=FLAGS.dataset_dir,
batch_size=clone_batch_size,
crop_size=[int(sz) for sz in FLAGS.train_crop_size],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
resize_factor=FLAGS.resize_factor,
min_scale_factor=FLAGS.min_scale_factor,
max_scale_factor=FLAGS.max_scale_factor,
scale_factor_step_size=FLAGS.scale_factor_step_size,
model_variant=FLAGS.model_variant,
num_readers=4,
is_training=True,
should_shuffle=True,
should_repeat=True)
# Create the global step on the device storing the variables.
with tf.device(config.variables_device()):
global_step = tf.train.get_or_create_global_step()
# Define the model and create clones.
model_fn = _build_deeplab
model_args = (dataset.get_one_shot_iterator(), {
common.OUTPUT_TYPE: dataset.num_of_classes
}, dataset.ignore_label)
clones = model_deploy.create_clones(config,
model_fn,
args=model_args)
# Gather update_ops from the first clone. These contain, for example,
# the updates for the batch_norm variables created by model_fn.
first_clone_scope = config.clone_scope(0)
update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS,
first_clone_scope)
# Gather initial summaries.
summaries = set(tf.get_collection(tf.GraphKeys.SUMMARIES))
# Add summaries for model variables.
for model_var in tf.model_variables():
summaries.add(tf.summary.histogram(model_var.op.name, model_var))
# Add summaries for images, labels, semantic predictions
if FLAGS.save_summaries_images:
summary_image = graph.get_tensor_by_name(
('%s/%s:0' % (first_clone_scope, common.IMAGE)).strip('/'))
summaries.add(
tf.summary.image('samples/%s' % common.IMAGE, summary_image))
first_clone_label = graph.get_tensor_by_name(
('%s/%s:0' % (first_clone_scope, common.LABEL)).strip('/'))
# Scale up summary image pixel values for better visualization.
pixel_scaling = max(1, 255 // dataset.num_of_classes)
summary_label = tf.cast(first_clone_label * pixel_scaling,
tf.uint8)
summaries.add(
tf.summary.image('samples/%s' % common.LABEL, summary_label))
first_clone_output = graph.get_tensor_by_name(
('%s/%s:0' %
(first_clone_scope, common.OUTPUT_TYPE)).strip('/'))
predictions = tf.expand_dims(tf.argmax(first_clone_output, 3), -1)
summary_predictions = tf.cast(predictions * pixel_scaling,
tf.uint8)
summaries.add(
tf.summary.image('samples/%s' % common.OUTPUT_TYPE,
summary_predictions))
# Add summaries for losses.
for loss in tf.get_collection(tf.GraphKeys.LOSSES, first_clone_scope):
summaries.add(tf.summary.scalar('losses/%s' % loss.op.name, loss))
# Build the optimizer based on the device specification.
with tf.device(config.optimizer_device()):
learning_rate = train_utils.get_model_learning_rate(
FLAGS.learning_policy,
FLAGS.base_learning_rate,
FLAGS.learning_rate_decay_step,
FLAGS.learning_rate_decay_factor,
FLAGS.training_number_of_steps,
FLAGS.learning_power,
FLAGS.slow_start_step,
FLAGS.slow_start_learning_rate,
decay_steps=FLAGS.decay_steps,
end_learning_rate=FLAGS.end_learning_rate)
summaries.add(tf.summary.scalar('learning_rate', learning_rate))
if FLAGS.optimizer == 'momentum':
optimizer = tf.train.MomentumOptimizer(learning_rate,
FLAGS.momentum)
elif FLAGS.optimizer == 'adam':
optimizer = tf.train.AdamOptimizer(
learning_rate=FLAGS.adam_learning_rate,
epsilon=FLAGS.adam_epsilon)
else:
raise ValueError('Unknown optimizer')
if FLAGS.quantize_delay_step >= 0:
if FLAGS.num_clones > 1:
raise ValueError(
'Quantization doesn\'t support multi-clone yet.')
contrib_quantize.create_training_graph(
quant_delay=FLAGS.quantize_delay_step)
startup_delay_steps = FLAGS.task * FLAGS.startup_delay_steps
with tf.device(config.variables_device()):
total_loss, grads_and_vars = model_deploy.optimize_clones(
clones, optimizer)
total_loss = tf.check_numerics(total_loss, 'Loss is inf or nan.')
summaries.add(tf.summary.scalar('total_loss', total_loss))
# Modify the gradients for biases and last layer variables.
last_layers = model.get_extra_layer_scopes(
FLAGS.last_layers_contain_logits_only)
grad_mult = train_utils.get_model_gradient_multipliers(
last_layers, FLAGS.last_layer_gradient_multiplier)
if grad_mult:
grads_and_vars = slim.learning.multiply_gradients(
grads_and_vars, grad_mult)
# Create gradient update op.
grad_updates = optimizer.apply_gradients(grads_and_vars,
global_step=global_step)
update_ops.append(grad_updates)
update_op = tf.group(*update_ops)
with tf.control_dependencies([update_op]):
train_tensor = tf.identity(total_loss, name='train_op')
# Add the summaries from the first clone. These contain the summaries
# created by model_fn and either optimize_clones() or _gather_clone_loss().
summaries |= set(
tf.get_collection(tf.GraphKeys.SUMMARIES, first_clone_scope))
# Merge all summaries together.
summary_op = tf.summary.merge(list(summaries))
# Soft placement allows placing on CPU ops without GPU implementation.
session_config = tf.ConfigProto(allow_soft_placement=True,
log_device_placement=False)
# Start the training.
profile_dir = FLAGS.profile_logdir
if profile_dir is not None:
tf.gfile.MakeDirs(profile_dir)
with contrib_tfprof.ProfileContext(enabled=profile_dir is not None,
profile_dir=profile_dir):
init_fn = None
if FLAGS.tf_initial_checkpoint:
init_fn = train_utils.get_model_init_fn(
FLAGS.train_logdir,
FLAGS.tf_initial_checkpoint,
FLAGS.initialize_last_layer,
last_layers,
ignore_missing_vars=True)
slim.learning.train(train_tensor,
logdir=FLAGS.train_logdir,
log_every_n_steps=FLAGS.log_steps,
master=FLAGS.master,
number_of_steps=FLAGS.training_number_of_steps,
is_chief=(FLAGS.task == 0),
session_config=session_config,
startup_delay_steps=startup_delay_steps,
init_fn=init_fn,
summary_op=summary_op,
save_summaries_secs=FLAGS.save_summaries_secs,
save_interval_secs=FLAGS.save_interval_secs)
def main(unused_argv):
# Check model parameters
check_model_conflict()
data_inforamtion = data_generator._DATASETS_INFORMATION[FLAGS.dataset_name]
tf.logging.set_verbosity(tf.logging.INFO)
tf.gfile.MakeDirs(FLAGS.train_logdir)
for split in FLAGS.train_split:
tf.logging.info('Training on %s set', split)
path = FLAGS.train_logdir
parameters_dict = vars(FLAGS)
with open(os.path.join(path, 'json.txt'), 'w', encoding='utf-8') as f:
json.dump(parameters_dict, f, indent=3)
with open(os.path.join(path, 'logging.txt'), 'w') as f:
for key in parameters_dict:
f.write("{}: {}".format(str(key), str(parameters_dict[key])))
f.write("\n")
f.write("\nStart time: {}".format(
time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())))
f.write("\n")
graph = tf.Graph()
with graph.as_default():
with tf.device(
tf.train.replica_device_setter(ps_tasks=FLAGS.num_ps_tasks)):
assert FLAGS.batch_size % FLAGS.num_clones == 0, (
'Training batch size not divisble by number of clones (GPUs).')
clone_batch_size = FLAGS.batch_size // FLAGS.num_clones
if FLAGS.dataset_name == '2019_ISBI_CHAOS_MR_T1' or FLAGS.dataset_name == '2019_ISBI_CHAOS_MR_T2':
min_resize_value = data_inforamtion.height
max_resize_value = data_inforamtion.height
else:
if FLAGS.min_resize_value is not None:
min_resize_value = FLAGS.min_resize_value
else:
min_resize_value = data_inforamtion.height
if FLAGS.max_resize_value is not None:
max_resize_value = FLAGS.max_resize_value
else:
max_resize_value = data_inforamtion.height
train_generator = data_generator.Dataset(
dataset_name=FLAGS.dataset_name,
split_name=FLAGS.train_split,
guidance_type=FLAGS.guidance_type,
batch_size=clone_batch_size,
pre_crop_flag=FLAGS.pre_crop_flag,
mt_class=FLAGS.mt_output_node,
crop_size=data_inforamtion.train["train_crop_size"],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
resize_factor=FLAGS.resize_factor,
min_scale_factor=FLAGS.min_scale_factor,
max_scale_factor=FLAGS.max_scale_factor,
scale_factor_step_size=FLAGS.scale_factor_step_size,
num_readers=2,
is_training=True,
shuffle_data=True,
repeat_data=True,
prior_num_slice=FLAGS.prior_num_slice,
prior_num_subject=FLAGS.prior_num_subject,
seq_length=FLAGS.seq_length,
seq_type="bidirection",
z_loss_name=FLAGS.z_loss_name,
)
if "val" not in FLAGS.train_split:
val_generator = data_generator.Dataset(
dataset_name=FLAGS.dataset_name,
split_name=["val"],
guidance_type=FLAGS.guidance_type,
batch_size=1,
mt_class=FLAGS.mt_output_node,
crop_size=[
data_inforamtion.height, data_inforamtion.width
],
min_resize_value=FLAGS.min_resize_value,
max_resize_value=FLAGS.max_resize_value,
num_readers=2,
is_training=False,
shuffle_data=False,
repeat_data=True,
prior_num_slice=FLAGS.prior_num_slice,
prior_num_subject=FLAGS.prior_num_subject,
seq_length=FLAGS.seq_length,
seq_type="bidirection",
z_loss_name=FLAGS.z_loss_name,
)
model_options = common.ModelOptions(
outputs_to_num_classes=train_generator.num_of_classes,
crop_size=data_inforamtion.train["train_crop_size"],
output_stride=FLAGS.output_stride)
steps = tf.compat.v1.placeholder(tf.int32, shape=[])
dataset1 = train_generator.get_dataset()
iter1 = dataset1.make_one_shot_iterator()
train_samples = iter1.get_next()
train_tensor, summary_op = _train_pgn_model(
train_samples, train_generator.num_of_classes, model_options,
train_generator.ignore_label)
if "val" not in FLAGS.train_split:
dataset2 = val_generator.get_dataset()
iter2 = dataset2.make_one_shot_iterator()
val_samples = iter2.get_next()
val_tensor, _ = _val_pgn_model(val_samples,
val_generator.num_of_classes,
model_options,
val_generator.ignore_label,
steps)
# Soft placement allows placing on CPU ops without GPU implementation.
session_config = tf.ConfigProto(allow_soft_placement=True,
log_device_placement=False)
init_fn = None
if FLAGS.tf_initial_checkpoint:
init_fn = train_utils.get_model_init_fn(
train_logdir=FLAGS.train_logdir,
tf_initial_checkpoint=FLAGS.tf_initial_checkpoint,
initialize_first_layer=True,
initialize_last_layer=FLAGS.initialize_last_layer,
ignore_missing_vars=True)
scaffold = tf.train.Scaffold(
init_fn=init_fn,
summary_op=summary_op,
)
stop_hook = tf.train.StopAtStepHook(FLAGS.training_number_of_steps)
saver = tf.train.Saver()
best_dice = 0
with tf.train.MonitoredTrainingSession(
master=FLAGS.master,
is_chief=(FLAGS.task == 0),
config=session_config,
scaffold=scaffold,
checkpoint_dir=FLAGS.train_logdir,
log_step_count_steps=FLAGS.log_steps,
save_summaries_steps=20,
save_checkpoint_steps=FLAGS.save_checkpoint_steps,
hooks=[stop_hook]) as sess:
# step=0
total_val_loss, total_val_steps = [], []
best_model_performance = 0.0
while not sess.should_stop():
_, global_step = sess.run(
[train_tensor,
tf.train.get_global_step()])
if "val" not in FLAGS.train_split:
if global_step % FLAGS.validation_steps == 0:
cm_total = 0
for j in range(
val_generator.splits_to_sizes["val"]):
cm_total += sess.run(val_tensor,
feed_dict={steps: j})
mean_dice_score, _ = metrics.compute_mean_dsc(
total_cm=cm_total)
total_val_loss.append(mean_dice_score)
total_val_steps.append(global_step)
plt.legend(["validation loss"])
plt.xlabel("global step")
plt.ylabel("loss")
plt.plot(total_val_steps, total_val_loss, "bo-")
plt.grid(True)
plt.savefig(FLAGS.train_logdir + "/losses.png")
if mean_dice_score > best_dice:
best_dice = mean_dice_score
saver.save(
get_session(sess),
os.path.join(FLAGS.train_logdir,
'model.ckpt-best'))
# saver.save(get_session(sess), os.path.join(FLAGS.train_logdir, 'model.ckpt-best-%d' %global_step))
txt = 20 * ">" + " saving best mdoel model.ckpt-best-%d with DSC: %f" % (
global_step, best_dice)
print(txt)
with open(os.path.join(path, 'logging.txt'),
'a') as f:
f.write(txt)
f.write("\n")
with open(os.path.join(path, 'logging.txt'), 'a') as f:
f.write("\nEnd time: {}".format(
time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())))
f.write("\n")
