Python object_detection.meta_architectures.ssd_meta_arch.SSDMetaArch() Examples
The following are 30
code examples of object_detection.meta_architectures.ssd_meta_arch.SSDMetaArch().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
object_detection.meta_architectures.ssd_meta_arch
, or try the search function
.
.
Example #1
| Source File: ssd_meta_arch_test.py From vehicle_counting_tensorflow with MIT License | 6 votes |
|
def _create_model(self,
apply_hard_mining=True,
normalize_loc_loss_by_codesize=False,
add_background_class=True,
random_example_sampling=False,
weight_regression_loss_by_score=False,
use_expected_classification_loss_under_sampling=False,
min_num_negative_samples=1,
desired_negative_sampling_ratio=3,
use_keras=False,
predict_mask=False,
use_static_shapes=False,
nms_max_size_per_class=5):
return super(SsdMetaArchTest, self)._create_model(
model_fn=ssd_meta_arch.SSDMetaArch,
apply_hard_mining=apply_hard_mining,
normalize_loc_loss_by_codesize=normalize_loc_loss_by_codesize,
add_background_class=add_background_class,
random_example_sampling=random_example_sampling,
weight_regression_loss_by_score=weight_regression_loss_by_score,
use_expected_classification_loss_under_sampling=
use_expected_classification_loss_under_sampling,
min_num_negative_samples=min_num_negative_samples,
desired_negative_sampling_ratio=desired_negative_sampling_ratio,
use_keras=use_keras,
predict_mask=predict_mask,
use_static_shapes=use_static_shapes,
nms_max_size_per_class=nms_max_size_per_class)
Example #2
| Source File: ssd_meta_arch_test.py From multilabel-image-classification-tensorflow with MIT License | 5 votes |
|
def _create_model(self,
apply_hard_mining=True,
normalize_loc_loss_by_codesize=False,
add_background_class=True,
random_example_sampling=False,
weight_regression_loss_by_score=False,
use_expected_classification_loss_under_sampling=False,
min_num_negative_samples=1,
desired_negative_sampling_ratio=3,
use_keras=False,
predict_mask=False,
use_static_shapes=False,
nms_max_size_per_class=5):
return super(SsdMetaArchTest, self)._create_model(
model_fn=ssd_meta_arch.SSDMetaArch,
apply_hard_mining=apply_hard_mining,
normalize_loc_loss_by_codesize=normalize_loc_loss_by_codesize,
add_background_class=add_background_class,
random_example_sampling=random_example_sampling,
weight_regression_loss_by_score=weight_regression_loss_by_score,
use_expected_classification_loss_under_sampling=
use_expected_classification_loss_under_sampling,
min_num_negative_samples=min_num_negative_samples,
desired_negative_sampling_ratio=desired_negative_sampling_ratio,
use_keras=use_keras,
predict_mask=predict_mask,
use_static_shapes=use_static_shapes,
nms_max_size_per_class=nms_max_size_per_class)
Example #3
| Source File: model_builder_test.py From Live-feed-object-device-identification-using-Tensorflow-and-OpenCV with Apache License 2.0 | 5 votes |
|
def test_create_ssd_models_from_config(self):
model_proto = self.create_default_ssd_model_proto()
ssd_feature_extractor_map = {}
ssd_feature_extractor_map.update(
model_builder.SSD_FEATURE_EXTRACTOR_CLASS_MAP)
ssd_feature_extractor_map.update(
model_builder.SSD_KERAS_FEATURE_EXTRACTOR_CLASS_MAP)
for extractor_type, extractor_class in ssd_feature_extractor_map.items():
model_proto.ssd.feature_extractor.type = extractor_type
model = model_builder.build(model_proto, is_training=True)
self.assertIsInstance(model, ssd_meta_arch.SSDMetaArch)
self.assertIsInstance(model._feature_extractor, extractor_class)
Example #4
| Source File: model_builder_test.py From Live-feed-object-device-identification-using-Tensorflow-and-OpenCV with Apache License 2.0 | 5 votes |
|
def test_create_ssd_models_from_config(self):
model_proto = self.create_default_ssd_model_proto()
ssd_feature_extractor_map = {}
ssd_feature_extractor_map.update(
model_builder.SSD_FEATURE_EXTRACTOR_CLASS_MAP)
ssd_feature_extractor_map.update(
model_builder.SSD_KERAS_FEATURE_EXTRACTOR_CLASS_MAP)
for extractor_type, extractor_class in ssd_feature_extractor_map.items():
model_proto.ssd.feature_extractor.type = extractor_type
model = model_builder.build(model_proto, is_training=True)
self.assertIsInstance(model, ssd_meta_arch.SSDMetaArch)
self.assertIsInstance(model._feature_extractor, extractor_class)
Example #5
| Source File: ssd_meta_arch_test.py From MAX-Object-Detector with Apache License 2.0 | 5 votes |
|
def _create_model(
self,
apply_hard_mining=True,
normalize_loc_loss_by_codesize=False,
add_background_class=True,
random_example_sampling=False,
expected_loss_weights=model_pb2.DetectionModel().ssd.loss.NONE,
min_num_negative_samples=1,
desired_negative_sampling_ratio=3,
use_keras=False,
predict_mask=False,
use_static_shapes=False,
nms_max_size_per_class=5):
return super(SsdMetaArchTest, self)._create_model(
model_fn=ssd_meta_arch.SSDMetaArch,
apply_hard_mining=apply_hard_mining,
normalize_loc_loss_by_codesize=normalize_loc_loss_by_codesize,
add_background_class=add_background_class,
random_example_sampling=random_example_sampling,
expected_loss_weights=expected_loss_weights,
min_num_negative_samples=min_num_negative_samples,
desired_negative_sampling_ratio=desired_negative_sampling_ratio,
use_keras=use_keras,
predict_mask=predict_mask,
use_static_shapes=use_static_shapes,
nms_max_size_per_class=nms_max_size_per_class)
Example #6
| Source File: ssd_meta_arch_test.py From g-tensorflow-models with Apache License 2.0 | 5 votes |
|
def _create_model(
self,
apply_hard_mining=True,
normalize_loc_loss_by_codesize=False,
add_background_class=True,
random_example_sampling=False,
expected_loss_weights=model_pb2.DetectionModel().ssd.loss.NONE,
min_num_negative_samples=1,
desired_negative_sampling_ratio=3,
use_keras=False,
predict_mask=False,
use_static_shapes=False,
nms_max_size_per_class=5):
return super(SsdMetaArchTest, self)._create_model(
model_fn=ssd_meta_arch.SSDMetaArch,
apply_hard_mining=apply_hard_mining,
normalize_loc_loss_by_codesize=normalize_loc_loss_by_codesize,
add_background_class=add_background_class,
random_example_sampling=random_example_sampling,
expected_loss_weights=expected_loss_weights,
min_num_negative_samples=min_num_negative_samples,
desired_negative_sampling_ratio=desired_negative_sampling_ratio,
use_keras=use_keras,
predict_mask=predict_mask,
use_static_shapes=use_static_shapes,
nms_max_size_per_class=nms_max_size_per_class)
Example #7
| Source File: ssd_meta_arch_test.py From models with Apache License 2.0 | 5 votes |
|
def _create_model(
self,
apply_hard_mining=True,
normalize_loc_loss_by_codesize=False,
add_background_class=True,
random_example_sampling=False,
expected_loss_weights=model_pb2.DetectionModel().ssd.loss.NONE,
min_num_negative_samples=1,
desired_negative_sampling_ratio=3,
predict_mask=False,
use_static_shapes=False,
nms_max_size_per_class=5,
calibration_mapping_value=None,
return_raw_detections_during_predict=False):
return super(SsdMetaArchTest, self)._create_model(
model_fn=ssd_meta_arch.SSDMetaArch,
apply_hard_mining=apply_hard_mining,
normalize_loc_loss_by_codesize=normalize_loc_loss_by_codesize,
add_background_class=add_background_class,
random_example_sampling=random_example_sampling,
expected_loss_weights=expected_loss_weights,
min_num_negative_samples=min_num_negative_samples,
desired_negative_sampling_ratio=desired_negative_sampling_ratio,
predict_mask=predict_mask,
use_static_shapes=use_static_shapes,
nms_max_size_per_class=nms_max_size_per_class,
calibration_mapping_value=calibration_mapping_value,
return_raw_detections_during_predict=(
return_raw_detections_during_predict))
Example #8
| Source File: model_builder_test.py From models with Apache License 2.0 | 5 votes |
|
def test_create_ssd_models_from_config(self):
model_proto = self.create_default_ssd_model_proto()
for extractor_type, extractor_class in self.ssd_feature_extractors().items(
):
model_proto.ssd.feature_extractor.type = extractor_type
model = model_builder.build(model_proto, is_training=True)
self.assertIsInstance(model, ssd_meta_arch.SSDMetaArch)
self.assertIsInstance(model._feature_extractor, extractor_class)
Example #9
| Source File: ssd_meta_arch_test.py From mtl-ssl with Apache License 2.0 | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #10
| Source File: model_builder.py From object_detection_with_tensorflow with MIT License | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #11
| Source File: ssd_meta_arch_test.py From object_detection_with_tensorflow with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #12
| Source File: model_builder.py From object_detection_with_tensorflow with MIT License | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #13
| Source File: ssd_meta_arch_test.py From object_detection_with_tensorflow with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #14
| Source File: model_builder.py From Elphas with Apache License 2.0 | 4 votes |
|
def _build_ssd_model(ssd_config, is_training, add_summaries):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
add_summaries: Whether to add tf summaries in the model.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
encode_background_as_zeros = ssd_config.encode_background_as_zeros
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
encode_background_as_zeros,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner,
add_summaries=add_summaries)
Example #15
| Source File: ssd_meta_arch_test.py From Elphas with Apache License 2.0 | 4 votes |
|
def _create_model(self, apply_hard_mining=True):
is_training = False
num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
encode_background_as_zeros = False
def image_resizer_fn(image):
return [tf.identity(image), tf.shape(image)]
classification_loss = losses.WeightedSigmoidClassificationLoss()
localization_loss = losses.WeightedSmoothL1LocalizationLoss()
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
hard_example_miner = None
if apply_hard_mining:
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
code_size = 4
model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
encode_background_as_zeros, image_resizer_fn, non_max_suppression_fn,
tf.identity, classification_loss, localization_loss,
classification_loss_weight, localization_loss_weight,
normalize_loss_by_num_matches, hard_example_miner, add_summaries=False)
return model, num_classes, mock_anchor_generator.num_anchors(), code_size
Example #16
| Source File: model_builder.py From MBMD with MIT License | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #17
| Source File: ssd_meta_arch_test.py From MBMD with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #18
| Source File: model_builder.py From object_detection_kitti with Apache License 2.0 | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #19
| Source File: ssd_meta_arch_test.py From object_detection_kitti with Apache License 2.0 | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #20
| Source File: model_builder.py From mtl-ssl with Apache License 2.0 | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training and ssd_config.feature_extractor.trainable)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training and ssd_config.box_predictor.trainable,
num_classes)
anchor_generator = anchor_generator_builder.build(ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #21
| Source File: ssd_meta_arch_test.py From hands-detection with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #22
| Source File: model_builder.py From yolo_v2 with Apache License 2.0 | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
common_kwargs = {
'is_training':is_training,
'anchor_generator':anchor_generator,
'box_predictor':ssd_box_predictor,
'box_coder':box_coder,
'feature_extractor':feature_extractor,
'matcher':matcher,
'region_similarity_calculator':region_similarity_calculator,
'image_resizer_fn':image_resizer_fn,
'non_max_suppression_fn':non_max_suppression_fn,
'score_conversion_fn':score_conversion_fn,
'classification_loss':classification_loss,
'localization_loss':localization_loss,
'classification_loss_weight':classification_weight,
'localization_loss_weight':localization_weight,
'normalize_loss_by_num_matches':normalize_loss_by_num_matches,
'hard_example_miner':hard_example_miner}
if isinstance(anchor_generator, yolo_grid_anchor_generator.YoloGridAnchorGenerator):
return yolo_meta_arch.YOLOMetaArch(**common_kwargs)
else:
return ssd_meta_arch.SSDMetaArch(**common_kwargs)
Example #23
| Source File: model_builder.py From DOTA_models with Apache License 2.0 | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #24
| Source File: ssd_meta_arch_test.py From object_detector_app with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #25
| Source File: model_builder.py From object_detector_app with MIT License | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #26
| Source File: ssd_meta_arch_test.py From ros_people_object_detection_tensorflow with Apache License 2.0 | 4 votes |
|
def _create_model(self, apply_hard_mining=True,
normalize_loc_loss_by_codesize=False):
is_training = False
num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
encode_background_as_zeros = False
def image_resizer_fn(image):
return [tf.identity(image), tf.shape(image)]
classification_loss = losses.WeightedSigmoidClassificationLoss()
localization_loss = losses.WeightedSmoothL1LocalizationLoss()
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
negative_class_weight = 1.0
normalize_loss_by_num_matches = False
hard_example_miner = None
if apply_hard_mining:
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
code_size = 4
model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
encode_background_as_zeros, negative_class_weight, image_resizer_fn,
non_max_suppression_fn, tf.identity, classification_loss,
localization_loss, classification_loss_weight, localization_loss_weight,
normalize_loss_by_num_matches, hard_example_miner, add_summaries=False,
normalize_loc_loss_by_codesize=normalize_loc_loss_by_codesize)
return model, num_classes, mock_anchor_generator.num_anchors(), code_size
Example #27
| Source File: ssd_meta_arch_test.py From garbage-object-detection-tensorflow with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #28
| Source File: model_builder.py From garbage-object-detection-tensorflow with MIT License | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
Example #29
| Source File: ssd_meta_arch_test.py From HereIsWally with MIT License | 4 votes |
|
def setUp(self):
"""Set up mock SSD model.
Here we set up a simple mock SSD model that will always predict 4
detections that happen to always be exactly the anchors that are set up
in the above MockAnchorGenerator. Because we let max_detections=5,
we will also always end up with an extra padded row in the detection
results.
"""
is_training = False
self._num_classes = 1
mock_anchor_generator = MockAnchorGenerator2x2()
mock_box_predictor = test_utils.MockBoxPredictor(
is_training, self._num_classes)
mock_box_coder = test_utils.MockBoxCoder()
fake_feature_extractor = FakeSSDFeatureExtractor()
mock_matcher = test_utils.MockMatcher()
region_similarity_calculator = sim_calc.IouSimilarity()
def image_resizer_fn(image):
return tf.identity(image)
classification_loss = losses.WeightedSigmoidClassificationLoss(
anchorwise_output=True)
localization_loss = losses.WeightedSmoothL1LocalizationLoss(
anchorwise_output=True)
non_max_suppression_fn = functools.partial(
post_processing.batch_multiclass_non_max_suppression,
score_thresh=-20.0,
iou_thresh=1.0,
max_size_per_class=5,
max_total_size=5)
classification_loss_weight = 1.0
localization_loss_weight = 1.0
normalize_loss_by_num_matches = False
# This hard example miner is expected to be a no-op.
hard_example_miner = losses.HardExampleMiner(
num_hard_examples=None,
iou_threshold=1.0)
self._num_anchors = 4
self._code_size = 4
self._model = ssd_meta_arch.SSDMetaArch(
is_training, mock_anchor_generator, mock_box_predictor, mock_box_coder,
fake_feature_extractor, mock_matcher, region_similarity_calculator,
image_resizer_fn, non_max_suppression_fn, tf.identity,
classification_loss, localization_loss, classification_loss_weight,
localization_loss_weight, normalize_loss_by_num_matches,
hard_example_miner)
Example #30
| Source File: model_builder.py From HereIsWally with MIT License | 4 votes |
|
def _build_ssd_model(ssd_config, is_training):
"""Builds an SSD detection model based on the model config.
Args:
ssd_config: A ssd.proto object containing the config for the desired
SSDMetaArch.
is_training: True if this model is being built for training purposes.
Returns:
SSDMetaArch based on the config.
Raises:
ValueError: If ssd_config.type is not recognized (i.e. not registered in
model_class_map).
"""
num_classes = ssd_config.num_classes
# Feature extractor
feature_extractor = _build_ssd_feature_extractor(ssd_config.feature_extractor,
is_training)
box_coder = box_coder_builder.build(ssd_config.box_coder)
matcher = matcher_builder.build(ssd_config.matcher)
region_similarity_calculator = sim_calc.build(
ssd_config.similarity_calculator)
ssd_box_predictor = box_predictor_builder.build(hyperparams_builder.build,
ssd_config.box_predictor,
is_training, num_classes)
anchor_generator = anchor_generator_builder.build(
ssd_config.anchor_generator)
image_resizer_fn = image_resizer_builder.build(ssd_config.image_resizer)
non_max_suppression_fn, score_conversion_fn = post_processing_builder.build(
ssd_config.post_processing)
(classification_loss, localization_loss, classification_weight,
localization_weight,
hard_example_miner) = losses_builder.build(ssd_config.loss)
normalize_loss_by_num_matches = ssd_config.normalize_loss_by_num_matches
return ssd_meta_arch.SSDMetaArch(
is_training,
anchor_generator,
ssd_box_predictor,
box_coder,
feature_extractor,
matcher,
region_similarity_calculator,
image_resizer_fn,
non_max_suppression_fn,
score_conversion_fn,
classification_loss,
localization_loss,
classification_weight,
localization_weight,
normalize_loss_by_num_matches,
hard_example_miner)
