def variable_summaries(var, scope=""):
"""Attach a lot of summaries to a Tensor (for TensorBoard visualization)."""
with tf.name_scope(scope):
with tf.name_scope("summaries"):
mean = tf.reduce_mean(var)
tf.summary.scalar("mean", mean)
with tf.name_scope("stddev"):
stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))
tf.summary.scalar("stddev", stddev)
tf.summary.scalar("max", tf.reduce_max(var))
tf.summary.scalar("min", tf.reduce_min(var))
tf.summary.histogram("histogram", var)
def metric_fn(loss):
"""Evaluation metric Fn which runs on CPU."""
perplexity = tf.exp(tf.reduce_mean(loss))
return {
"eval/loss": tf.metrics.mean(loss),
"eval/perplexity": tf.metrics.mean(perplexity),
}
def _discriminator(self, encoder_output, features, labels, num_classes, num_units=256):
# A class discriminator
# Create attention over the encoder states
encoder_output_output = encoder_output["outputs"]
encoder_output_output_shape = encoder_output_output.get_shape()
encoder_output_att_values = encoder_output["attention_values"]
encoder_output_att_values_shape = encoder_output_att_values.get_shape()
if self.params["discriminator_reverse_grad"]:
encoder_output_output = reverse_grad(encoder_output_output)
encoder_output_output.set_shape(encoder_output_output_shape)
encoder_output_att_values = reverse_grad(encoder_output_att_values)
encoder_output_att_values.set_shape(encoder_output_att_values_shape)
attention_fn = AttentionLayerBahdanau(params={}, mode=self.mode)
_, attention_context = attention_fn(
query=tf.zeros_like(encoder_output["outputs"][:, 0, :]),
keys=encoder_output_output,
values=encoder_output_att_values,
values_length=encoder_output["attention_values_length"])
# Fully connected layer
fc1 = tf.contrib.layers.fully_connected(
inputs=attention_context,
num_outputs=num_units,
activation_fn=tf.nn.tanh,
scope="discriminator_fc")
# Create logits
logits = tf.contrib.layers.fully_connected(
inputs=fc1,
num_outputs=num_classes,
activation_fn=None,
scope="discriminator_softmax")
losses = tf.nn.sparse_softmax_cross_entropy_with_logits(
logits=logits, labels=labels["domain"])
mean_loss = tf.reduce_mean(losses, name="mean_loss")
return (mean_loss, fc1)
def _process_example(images, cls_targets, box_targets, num_positives,
source_ids, image_scales, boxes, is_crowds, areas,
classes):
"""Processes one batch of data."""
labels = {}
# Count num_positives in a batch.
num_positives_batch = tf.reduce_mean(num_positives)
labels['mean_num_positives'] = tf.reshape(
tf.tile(tf.expand_dims(num_positives_batch, 0), [
batch_size,
]), [batch_size, 1])
for level in range(params['min_level'], params['max_level'] + 1):
labels['cls_targets_%d' % level] = cls_targets[level]
labels['box_targets_%d' % level] = box_targets[level]
# Concatenate groundtruth annotations to a tensor.
groundtruth_data = tf.concat([boxes, is_crowds, areas, classes], axis=2)
labels['source_ids'] = source_ids
labels['groundtruth_data'] = groundtruth_data
labels['image_scales'] = image_scales
return images, labels
def __call__(self, params):
input_anchors = anchors.Anchors(params['min_level'],
params['max_level'],
params['num_scales'],
params['aspect_ratios'],
params['anchor_scale'],
params['image_size'])
anchor_labeler = anchors.AnchorLabeler(input_anchors,
params['num_classes'])
example_decoder = tf_example_decoder.TfExampleDecoder()
def get_dataset_for_mode(data_dir, is_training):
"""Return the location of input samples for a given mode."""
if is_training:
return '%s/coco_train2017_nocrowd-*' % data_dir
return '%s/coco_val2017-*' % data_dir
def _dataset_parser(value):
"""Parse data to a fixed dimension input image and learning targets."""
with tf.name_scope('parser'):
data = example_decoder.decode(value)
source_id = data['source_id']
image = data['image']
boxes = data['groundtruth_boxes']
classes = data['groundtruth_classes']
classes = tf.reshape(tf.cast(classes, dtype=tf.float32),
[-1, 1])
# the image normalization is identical to Cloud TPU ResNet-50
image = tf.image.convert_image_dtype(image, dtype=tf.float32)
image = _normalize_image(image)
if params['input_rand_hflip']:
image, boxes = preprocessor.random_horizontal_flip(
image, boxes=boxes)
image_original_shape = tf.shape(image)
image, _ = preprocessor.resize_to_range(
image,
min_dimension=params['image_size'],
max_dimension=params['image_size'])
image_scale = tf.to_float(
image_original_shape[0]) / tf.to_float(tf.shape(image)[0])
image, boxes = preprocessor.scale_boxes_to_pixel_coordinates(
image, boxes, keypoints=None)
image = tf.image.pad_to_bounding_box(image, 0, 0,
params['image_size'],
params['image_size'])
(cls_targets, box_targets,
num_positives) = anchor_labeler.label_anchors(boxes, classes)
source_id = tf.string_to_number(source_id, out_type=tf.float32)
row = (image, cls_targets, box_targets, num_positives,
source_id, image_scale)
return row
batch_size = params['batch_size']
data_file_pattern = get_dataset_for_mode(self._data_dir,
self._is_training)
dataset = tf.data.Dataset.list_files(data_file_pattern)
dataset = dataset.shuffle(buffer_size=1024)
if self._is_training:
dataset = dataset.repeat()
def prefetch_dataset(filename):
dataset = tf.data.TFRecordDataset(filename).prefetch(1)
return dataset
dataset = dataset.apply(
tf.contrib.data.parallel_interleave(prefetch_dataset,
cycle_length=32,
sloppy=True))
dataset = dataset.shuffle(20)
dataset = dataset.map(_dataset_parser, num_parallel_calls=64)
dataset = dataset.prefetch(batch_size)
dataset = dataset.apply(
tf.contrib.data.batch_and_drop_remainder(batch_size))
dataset = dataset.prefetch(1)
(images, cls_targets, box_targets, num_positives, source_ids,
image_scales) = dataset.make_one_shot_iterator().get_next()
labels = {}
# count num_positives in a batch
num_positives_batch = tf.reduce_mean(num_positives)
labels['mean_num_positives'] = tf.reshape(
tf.tile(tf.expand_dims(num_positives_batch, 0), [
batch_size,
]), [batch_size, 1])
for level in range(params['min_level'], params['max_level'] + 1):
labels['cls_targets_%d' % level] = cls_targets[level]
labels['box_targets_%d' % level] = box_targets[level]
labels['source_ids'] = source_ids
labels['image_scales'] = image_scales
return images, labels
