def test_return_batch_norm_params_with_notrain_when_train_is_false(self):
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
batch_norm {
decay: 0.7
center: false
scale: true
epsilon: 0.03
train: false
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope[_get_scope_key(slim.conv2d)]
self.assertEqual(conv_scope_arguments['normalizer_fn'],
slim.batch_norm)
batch_norm_params = scope[_get_scope_key(slim.batch_norm)]
self.assertAlmostEqual(batch_norm_params['decay'], 0.7)
self.assertAlmostEqual(batch_norm_params['epsilon'], 0.03)
self.assertFalse(batch_norm_params['center'])
self.assertTrue(batch_norm_params['scale'])
self.assertFalse(batch_norm_params['is_training'])
def _build_arg_scope_with_conv_hyperparams(self):
conv_hyperparams = hyperparams_pb2.Hyperparams()
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
"""
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams)
return hyperparams_builder.build(conv_hyperparams, is_training=True)
def _build_arg_scope_with_hyperparams(
self, op_type=hyperparams_pb2.Hyperparams.FC):
hyperparams = hyperparams_pb2.Hyperparams()
hyperparams_text_proto = """
activation: NONE
regularizer {
l2_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
"""
text_format.Merge(hyperparams_text_proto, hyperparams)
hyperparams.op = op_type
return hyperparams_builder.build(hyperparams, is_training=True)
def test_default_arg_scope_has_conv2d_transpose_op(self):
conv_hyperparams_text_proto = """
regularizer {
l1_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
self.assertTrue(_get_scope_key(slim.conv2d_transpose) in scope)
def _build_conv_arg_scope_no_batch_norm(self):
conv_hyperparams = hyperparams_pb2.Hyperparams()
conv_hyperparams_text_proto = """
activation: RELU_6
regularizer {
l2_regularizer {
}
}
initializer {
random_normal_initializer {
stddev: 0.01
mean: 0.0
}
}
"""
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams)
return hyperparams_builder.build(conv_hyperparams, is_training=True)
def test_explicit_fc_op_arg_scope_has_fully_connected_op(self):
conv_hyperparams_text_proto = """
op: FC
regularizer {
l1_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
self.assertTrue(_get_scope_key(slim.fully_connected) in scope)
def test_do_not_use_batch_norm_if_default(self):
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope[_get_scope_key(slim.conv2d)]
self.assertEqual(conv_scope_arguments['normalizer_fn'], None)
def test_use_relu_6_activation(self):
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
activation: RELU_6
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope[_get_scope_key(slim.conv2d)]
self.assertEqual(conv_scope_arguments['activation_fn'], tf.nn.relu6)
def test_separable_conv2d_and_conv2d_and_transpose_have_same_parameters(
self):
conv_hyperparams_text_proto = """
regularizer {
l1_regularizer {
}
}
initializer {
truncated_normal_initializer {
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
kwargs_1, kwargs_2, kwargs_3 = scope.values()
self.assertDictEqual(kwargs_1, kwargs_2)
self.assertDictEqual(kwargs_1, kwargs_3)
def test_return_l1_regularized_weights(self):
conv_hyperparams_text_proto = """
regularizer {
l1_regularizer {
weight: 0.5
}
}
initializer {
truncated_normal_initializer {
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope.values()[0]
regularizer = conv_scope_arguments['weights_regularizer']
weights = np.array([1., -1, 4., 2.])
with self.test_session() as sess:
result = sess.run(regularizer(tf.constant(weights)))
self.assertAllClose(np.abs(weights).sum() * 0.5, result)
def test_variance_in_range_with_random_normal_initializer(self):
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
}
}
initializer {
random_normal_initializer {
mean: 0.0
stddev: 0.8
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope[_get_scope_key(slim.conv2d)]
initializer = conv_scope_arguments['weights_initializer']
self._assert_variance_in_range(initializer,
shape=[100, 40],
variance=0.64,
tol=1e-1)
def test_variance_in_range_with_variance_scaling_initializer_uniform(self):
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
}
}
initializer {
variance_scaling_initializer {
factor: 2.0
mode: FAN_IN
uniform: true
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope[_get_scope_key(slim.conv2d)]
initializer = conv_scope_arguments['weights_initializer']
self._assert_variance_in_range(initializer,
shape=[100, 40],
variance=2. / 100.)
def test_return_l2_regularizer_weights(self):
conv_hyperparams_text_proto = """
regularizer {
l2_regularizer {
weight: 0.42
}
}
initializer {
truncated_normal_initializer {
}
}
"""
conv_hyperparams_proto = hyperparams_pb2.Hyperparams()
text_format.Merge(conv_hyperparams_text_proto, conv_hyperparams_proto)
scope_fn = hyperparams_builder.build(conv_hyperparams_proto,
is_training=True)
scope = scope_fn()
conv_scope_arguments = scope[_get_scope_key(slim.conv2d)]
regularizer = conv_scope_arguments['weights_regularizer']
weights = np.array([1., -1, 4., 2.])
with self.test_session() as sess:
result = sess.run(regularizer(tf.constant(weights)))
self.assertAllClose(np.power(weights, 2).sum() / 2.0 * 0.42, result)
def _build_faster_rcnn_model(frcnn_config, is_training, add_summaries):
"""Builds a Faster R-CNN or R-FCN detection model based on the model config.
Builds R-FCN model if the second_stage_box_predictor in the config is of type
`rfcn_box_predictor` else builds a Faster R-CNN model.
Args:
frcnn_config: A faster_rcnn.proto object containing the config for the
desired FasterRCNNMetaArch or RFCNMetaArch.
is_training: True if this model is being built for training purposes.
add_summaries: Whether to add tf summaries in the model.
Returns:
FasterRCNNMetaArch based on the config.
Raises:
ValueError: If frcnn_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = frcnn_config.num_classes
image_resizer_fn = image_resizer_builder.build(frcnn_config.image_resizer)
is_keras = (frcnn_config.feature_extractor.type in
FASTER_RCNN_KERAS_FEATURE_EXTRACTOR_CLASS_MAP)
if is_keras:
feature_extractor = _build_faster_rcnn_keras_feature_extractor(
frcnn_config.feature_extractor, is_training,
inplace_batchnorm_update=frcnn_config.inplace_batchnorm_update)
else:
feature_extractor = _build_faster_rcnn_feature_extractor(
frcnn_config.feature_extractor, is_training,
inplace_batchnorm_update=frcnn_config.inplace_batchnorm_update)
number_of_stages = frcnn_config.number_of_stages
first_stage_anchor_generator = anchor_generator_builder.build(
frcnn_config.first_stage_anchor_generator)
first_stage_target_assigner = target_assigner.create_target_assigner(
'FasterRCNN',
'proposal',
use_matmul_gather=frcnn_config.use_matmul_gather_in_matcher)
first_stage_atrous_rate = frcnn_config.first_stage_atrous_rate
if is_keras:
first_stage_box_predictor_arg_scope_fn = (
hyperparams_builder.KerasLayerHyperparams(
frcnn_config.first_stage_box_predictor_conv_hyperparams))
else:
first_stage_box_predictor_arg_scope_fn = hyperparams_builder.build(
frcnn_config.first_stage_box_predictor_conv_hyperparams, is_training)
first_stage_box_predictor_kernel_size = (
frcnn_config.first_stage_box_predictor_kernel_size)
first_stage_box_predictor_depth = frcnn_config.first_stage_box_predictor_depth
first_stage_minibatch_size = frcnn_config.first_stage_minibatch_size
use_static_shapes = frcnn_config.use_static_shapes and (
frcnn_config.use_static_shapes_for_eval or is_training)
first_stage_sampler = sampler.BalancedPositiveNegativeSampler(
positive_fraction=frcnn_config.first_stage_positive_balance_fraction,
is_static=(frcnn_config.use_static_balanced_label_sampler and
use_static_shapes))
first_stage_max_proposals = frcnn_config.first_stage_max_proposals
if (frcnn_config.first_stage_nms_iou_threshold < 0 or
frcnn_config.first_stage_nms_iou_threshold > 1.0):
raise ValueError('iou_threshold not in [0, 1.0].')
if (is_training and frcnn_config.second_stage_batch_size >
first_stage_max_proposals):
raise ValueError('second_stage_batch_size should be no greater than '
'first_stage_max_proposals.')
first_stage_non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=frcnn_config.first_stage_nms_score_threshold,
iou_thresh=frcnn_config.first_stage_nms_iou_threshold,
max_size_per_class=frcnn_config.first_stage_max_proposals,
max_total_size=frcnn_config.first_stage_max_proposals,
use_static_shapes=use_static_shapes)
first_stage_loc_loss_weight = (
frcnn_config.first_stage_localization_loss_weight)
first_stage_obj_loss_weight = frcnn_config.first_stage_objectness_loss_weight
initial_crop_size = frcnn_config.initial_crop_size
maxpool_kernel_size = frcnn_config.maxpool_kernel_size
maxpool_stride = frcnn_config.maxpool_stride
second_stage_target_assigner = target_assigner.create_target_assigner(
'FasterRCNN',
'detection',
use_matmul_gather=frcnn_config.use_matmul_gather_in_matcher)
if is_keras:
second_stage_box_predictor = box_predictor_builder.build_keras(
hyperparams_builder.KerasLayerHyperparams,
freeze_batchnorm=False,
inplace_batchnorm_update=False,
num_predictions_per_location_list=[1],
box_predictor_config=frcnn_config.second_stage_box_predictor,
is_training=is_training,
num_classes=num_classes)
else:
second_stage_box_predictor = box_predictor_builder.build(
hyperparams_builder.build,
frcnn_config.second_stage_box_predictor,
is_training=is_training,
num_classes=num_classes)
second_stage_batch_size = frcnn_config.second_stage_batch_size
second_stage_sampler = sampler.BalancedPositiveNegativeSampler(
positive_fraction=frcnn_config.second_stage_balance_fraction,
is_static=(frcnn_config.use_static_balanced_label_sampler and
use_static_shapes))
(second_stage_non_max_suppression_fn, second_stage_score_conversion_fn
) = post_processing_builder.build(frcnn_config.second_stage_post_processing)
second_stage_localization_loss_weight = (
frcnn_config.second_stage_localization_loss_weight)
second_stage_classification_loss = (
losses_builder.build_faster_rcnn_classification_loss(
frcnn_config.second_stage_classification_loss))
second_stage_classification_loss_weight = (
frcnn_config.second_stage_classification_loss_weight)
second_stage_mask_prediction_loss_weight = (
frcnn_config.second_stage_mask_prediction_loss_weight)
hard_example_miner = None
if frcnn_config.HasField('hard_example_miner'):
hard_example_miner = losses_builder.build_hard_example_miner(
frcnn_config.hard_example_miner,
second_stage_classification_loss_weight,
second_stage_localization_loss_weight)
crop_and_resize_fn = (
ops.matmul_crop_and_resize if frcnn_config.use_matmul_crop_and_resize
else ops.native_crop_and_resize)
clip_anchors_to_image = (
frcnn_config.clip_anchors_to_image)
common_kwargs = {
'is_training': is_training,
'num_classes': num_classes,
'image_resizer_fn': image_resizer_fn,
'feature_extractor': feature_extractor,
'number_of_stages': number_of_stages,
'first_stage_anchor_generator': first_stage_anchor_generator,
'first_stage_target_assigner': first_stage_target_assigner,
'first_stage_atrous_rate': first_stage_atrous_rate,
'first_stage_box_predictor_arg_scope_fn':
first_stage_box_predictor_arg_scope_fn,
'first_stage_box_predictor_kernel_size':
first_stage_box_predictor_kernel_size,
'first_stage_box_predictor_depth': first_stage_box_predictor_depth,
'first_stage_minibatch_size': first_stage_minibatch_size,
'first_stage_sampler': first_stage_sampler,
'first_stage_non_max_suppression_fn': first_stage_non_max_suppression_fn,
'first_stage_max_proposals': first_stage_max_proposals,
'first_stage_localization_loss_weight': first_stage_loc_loss_weight,
'first_stage_objectness_loss_weight': first_stage_obj_loss_weight,
'second_stage_target_assigner': second_stage_target_assigner,
'second_stage_batch_size': second_stage_batch_size,
'second_stage_sampler': second_stage_sampler,
'second_stage_non_max_suppression_fn':
second_stage_non_max_suppression_fn,
'second_stage_score_conversion_fn': second_stage_score_conversion_fn,
'second_stage_localization_loss_weight':
second_stage_localization_loss_weight,
'second_stage_classification_loss':
second_stage_classification_loss,
'second_stage_classification_loss_weight':
second_stage_classification_loss_weight,
'hard_example_miner': hard_example_miner,
'add_summaries': add_summaries,
'crop_and_resize_fn': crop_and_resize_fn,
'clip_anchors_to_image': clip_anchors_to_image,
'use_static_shapes': use_static_shapes,
'resize_masks': frcnn_config.resize_masks
}
if (isinstance(second_stage_box_predictor,
rfcn_box_predictor.RfcnBoxPredictor) or
isinstance(second_stage_box_predictor,
rfcn_keras_box_predictor.RfcnKerasBoxPredictor)):
return rfcn_meta_arch.RFCNMetaArch(
second_stage_rfcn_box_predictor=second_stage_box_predictor,
**common_kwargs)
else:
return faster_rcnn_meta_arch.FasterRCNNMetaArch(
initial_crop_size=initial_crop_size,
maxpool_kernel_size=maxpool_kernel_size,
maxpool_stride=maxpool_stride,
second_stage_mask_rcnn_box_predictor=second_stage_box_predictor,
second_stage_mask_prediction_loss_weight=(
second_stage_mask_prediction_loss_weight),
**common_kwargs)
def _build_ssd_feature_extractor(feature_extractor_config,
is_training,
freeze_batchnorm,
reuse_weights=None):
"""Builds a ssd_meta_arch.SSDFeatureExtractor based on config.
Args:
feature_extractor_config: A SSDFeatureExtractor proto config from ssd.proto.
is_training: True if this feature extractor is being built for training.
freeze_batchnorm: Whether to freeze batch norm parameters during
training or not. When training with a small batch size (e.g. 1), it is
desirable to freeze batch norm update and use pretrained batch norm
params.
reuse_weights: if the feature extractor should reuse weights.
Returns:
ssd_meta_arch.SSDFeatureExtractor based on config.
Raises:
ValueError: On invalid feature extractor type.
"""
feature_type = feature_extractor_config.type
is_keras_extractor = feature_type in SSD_KERAS_FEATURE_EXTRACTOR_CLASS_MAP
depth_multiplier = feature_extractor_config.depth_multiplier
min_depth = feature_extractor_config.min_depth
pad_to_multiple = feature_extractor_config.pad_to_multiple
use_explicit_padding = feature_extractor_config.use_explicit_padding
use_depthwise = feature_extractor_config.use_depthwise
if is_keras_extractor:
conv_hyperparams = hyperparams_builder.KerasLayerHyperparams(
feature_extractor_config.conv_hyperparams)
else:
conv_hyperparams = hyperparams_builder.build(
feature_extractor_config.conv_hyperparams, is_training)
override_base_feature_extractor_hyperparams = (
feature_extractor_config.override_base_feature_extractor_hyperparams)
if (feature_type not in SSD_FEATURE_EXTRACTOR_CLASS_MAP) and (
not is_keras_extractor):
raise ValueError('Unknown ssd feature_extractor: {}'.format(feature_type))
if is_keras_extractor:
feature_extractor_class = SSD_KERAS_FEATURE_EXTRACTOR_CLASS_MAP[
feature_type]
else:
feature_extractor_class = SSD_FEATURE_EXTRACTOR_CLASS_MAP[feature_type]
kwargs = {
'is_training':
is_training,
'depth_multiplier':
depth_multiplier,
'min_depth':
min_depth,
'pad_to_multiple':
pad_to_multiple,
'use_explicit_padding':
use_explicit_padding,
'use_depthwise':
use_depthwise,
'override_base_feature_extractor_hyperparams':
override_base_feature_extractor_hyperparams
}
if feature_extractor_config.HasField('replace_preprocessor_with_placeholder'):
kwargs.update({
'replace_preprocessor_with_placeholder':
feature_extractor_config.replace_preprocessor_with_placeholder
})
if feature_extractor_config.HasField('num_layers'):
kwargs.update({'num_layers': feature_extractor_config.num_layers})
if is_keras_extractor:
kwargs.update({
'conv_hyperparams': conv_hyperparams,
'inplace_batchnorm_update': False,
'freeze_batchnorm': freeze_batchnorm
})
else:
kwargs.update({
'conv_hyperparams_fn': conv_hyperparams,
'reuse_weights': reuse_weights,
})
if feature_extractor_config.HasField('fpn'):
kwargs.update({
'fpn_min_level':
feature_extractor_config.fpn.min_level,
'fpn_max_level':
feature_extractor_config.fpn.max_level,
'additional_layer_depth':
feature_extractor_config.fpn.additional_layer_depth,
})
return feature_extractor_class(**kwargs)
