def test_model_initializing(self, init_checkpoint_modules):
shared_backbone = ('segmentation_backbone' not in init_checkpoint_modules)
shared_decoder = ('segmentation_decoder' not in init_checkpoint_modules and
shared_backbone)
task_config = cfg.PanopticMaskRCNNTask(
model=cfg.PanopticMaskRCNN(
num_classes=2,
input_size=[640, 640, 3],
segmentation_model=segmentation_cfg.SemanticSegmentationModel(
decoder=decoder_cfg.Decoder(type='fpn')),
shared_backbone=shared_backbone,
shared_decoder=shared_decoder))
task = panoptic_maskrcnn.PanopticMaskRCNNTask(task_config)
model = task.build_model()
ckpt = tf.train.Checkpoint(**model.checkpoint_items)
ckpt_save_dir = self.create_tempdir().full_path
ckpt.save(os.path.join(ckpt_save_dir, 'ckpt'))
if (init_checkpoint_modules == ['all'] or
'backbone' in init_checkpoint_modules):
task._task_config.init_checkpoint = ckpt_save_dir
if ('segmentation_backbone' in init_checkpoint_modules or
'segmentation_decoder' in init_checkpoint_modules):
task._task_config.segmentation_init_checkpoint = ckpt_save_dir
task._task_config.init_checkpoint_modules = init_checkpoint_modules
task.initialize(model)
def test_aspp_decoder_creation(self, level, dilation_rates, num_filters):
"""Test creation of ASPP decoder."""
input_specs = {'1': tf.TensorShape([1, 128, 128, 3])}
network = decoders.ASPP(level=level,
dilation_rates=dilation_rates,
num_filters=num_filters,
use_sync_bn=True)
model_config = configs.semantic_segmentation.SemanticSegmentationModel(
)
model_config.num_classes = 10
model_config.input_size = [None, None, 3]
model_config.decoder = decoders_cfg.Decoder(
type='aspp',
aspp=decoders_cfg.ASPP(level=level,
dilation_rates=dilation_rates,
num_filters=num_filters))
factory_network = factory.build_decoder(input_specs=input_specs,
model_config=model_config)
network_config = network.get_config()
factory_network_config = factory_network.get_config()
# Due to calling `super().get_config()` in aspp layer, everything but the
# the name of two layer instances are the same, so we force equal name so it
# will not give false alarm.
factory_network_config['name'] = network_config['name']
self.assertEqual(network_config, factory_network_config)
def test_deeplabv3_builder(self, backbone_type, input_size, weight_decay):
num_classes = 21
input_specs = tf.keras.layers.InputSpec(
shape=[None, input_size[0], input_size[1], 3])
model_config = semantic_segmentation_cfg.SemanticSegmentationModel(
num_classes=num_classes,
backbone=backbones.Backbone(type=backbone_type,
mobilenet=backbones.MobileNet(
model_id='MobileNetV2',
output_stride=16)),
decoder=decoders.Decoder(type='aspp',
aspp=decoders.ASPP(level=4,
num_filters=256,
dilation_rates=[],
spp_layer_version='v1',
output_tensor=True)),
head=semantic_segmentation_cfg.SegmentationHead(
level=4,
low_level=2,
num_convs=1,
upsample_factor=2,
use_depthwise_convolution=True))
l2_regularizer = (tf.keras.regularizers.l2(weight_decay)
if weight_decay else None)
model = factory.build_segmentation_model(input_specs=input_specs,
model_config=model_config,
l2_regularizer=l2_regularizer)
quantization_config = common.Quantization()
_ = qat_factory.build_qat_segmentation_model(
model=model,
quantization=quantization_config,
input_specs=input_specs)
def test_nasfpn_decoder_creation(self, num_filters, num_repeats,
use_separable_conv):
"""Test creation of NASFPN decoder."""
min_level = 3
max_level = 7
input_specs = {}
for level in range(min_level, max_level):
input_specs[str(level)] = tf.TensorShape(
[1, 128 // (2**level), 128 // (2**level), 3])
network = decoders.NASFPN(input_specs=input_specs,
num_filters=num_filters,
num_repeats=num_repeats,
use_separable_conv=use_separable_conv,
use_sync_bn=True)
model_config = configs.retinanet.RetinaNet()
model_config.min_level = min_level
model_config.max_level = max_level
model_config.num_classes = 10
model_config.input_size = [None, None, 3]
model_config.decoder = decoders_cfg.Decoder(
type='nasfpn',
nasfpn=decoders_cfg.NASFPN(num_filters=num_filters,
num_repeats=num_repeats,
use_separable_conv=use_separable_conv))
factory_network = factory.build_decoder(input_specs=input_specs,
model_config=model_config)
network_config = network.get_config()
factory_network_config = factory_network.get_config()
self.assertEqual(network_config, factory_network_config)
def test_aspp_decoder_creation(self, level, dilation_rates, num_filters):
"""Test creation of ASPP decoder."""
input_specs = {'1': tf.TensorShape([1, 128, 128, 3])}
network = decoders.ASPP(
level=level,
dilation_rates=dilation_rates,
num_filters=num_filters,
use_sync_bn=True)
model_config = configs.semantic_segmentation.SemanticSegmentationModel()
model_config.num_classes = 10
model_config.input_size = [None, None, 3]
model_config.decoder = decoders_cfg.Decoder(
type='aspp',
aspp=decoders_cfg.ASPP(
level=level, dilation_rates=dilation_rates,
num_filters=num_filters))
factory_network = factory.build_decoder(
input_specs=input_specs, model_config=model_config)
network_config = network.get_config()
factory_network_config = factory_network.get_config()
self.assertEqual(network_config, factory_network_config)
class Submodel(hyperparams.Config):
name: str = 'foo'
num_classes: int = 0
min_level: int = 3 # only for FPN or NASFPN
max_level: int = 6 # only for FPN or NASFPN
head: hyperparams.Config = SegmentationHead()
decoder: decoders.Decoder = decoders.Decoder(type='identity')
init_checkpoint_modules: str = None
class YoloModel(hyperparams.Config):
num_classes: int = 0
input_size: List[int] = dataclasses.field(default_factory=list)
min_level: int = 3 # only for FPN or NASFPN
max_level: int = 6 # only for FPN or NASFPN
head: hyperparams.Config = YoloHead()
backbone: backbones.Backbone = backbones.Backbone(
type='resnet', resnet=backbones.ResNet())
decoder: decoders.Decoder = decoders.Decoder(type='identity')
norm_activation: common.NormActivation = common.NormActivation()
class SemanticSegmentationModel(hyperparams.Config):
"""Semantic segmentation model config."""
num_classes: int = 0
input_size: List[int] = dataclasses.field(default_factory=list)
min_level: int = 3
max_level: int = 6
head: SegmentationHead = SegmentationHead()
backbone: backbones.Backbone = backbones.Backbone(
type='resnet', resnet=backbones.ResNet())
decoder: decoders.Decoder = decoders.Decoder(type='identity')
norm_activation: common.NormActivation = common.NormActivation()
class BASNetModel(hyperparams.Config):
"""BASNet model config."""
num_classes: int = 0
input_size: List[int] = dataclasses.field(default_factory=list)
#min_level: int = 3
#max_level: int = 6
#head: BASNetHead = BASNetHead()
backbone: backbones.Backbone = backbones.Backbone(
type='basnet_en', basnet_en=backbones.BASNet_En())
decoder: decoders.Decoder = decoders.Decoder(type='basnet_de')
norm_activation: common.NormActivation = common.NormActivation()
def test_identity_decoder_creation(self):
"""Test creation of identity decoder."""
model_config = configs.retinanet.RetinaNet()
model_config.num_classes = 2
model_config.input_size = [None, None, 3]
model_config.decoder = decoders_cfg.Decoder(
type='identity', identity=decoders_cfg.Identity())
factory_network = factory.build_decoder(input_specs=None,
model_config=model_config)
self.assertIsNone(factory_network)
class RetinaNet(hyperparams.Config):
num_classes: int = 0
input_size: List[int] = dataclasses.field(default_factory=list)
min_level: int = 3
max_level: int = 7
anchor: Anchor = Anchor()
backbone: backbones.Backbone = backbones.Backbone(
type='resnet', resnet=backbones.ResNet())
decoder: decoders.Decoder = decoders.Decoder(type='fpn',
fpn=decoders.FPN())
head: RetinaNetHead = RetinaNetHead()
detection_generator: DetectionGenerator = DetectionGenerator()
norm_activation: common.NormActivation = common.NormActivation()
class MaskRCNN(hyperparams.Config):
num_classes: int = 0
input_size: List[int] = dataclasses.field(default_factory=list)
min_level: int = 2
max_level: int = 6
anchor: Anchor = Anchor()
include_mask: bool = True
backbone: backbones.Backbone = backbones.Backbone(
type='resnet', resnet=backbones.ResNet())
decoder: decoders.Decoder = decoders.Decoder(type='fpn',
fpn=decoders.FPN())
rpn_head: RPNHead = RPNHead()
detection_head: DetectionHead = DetectionHead()
roi_generator: ROIGenerator = ROIGenerator()
roi_sampler: ROISampler = ROISampler()
roi_aligner: ROIAligner = ROIAligner()
detection_generator: DetectionGenerator = DetectionGenerator()
mask_head: Optional[MaskHead] = MaskHead()
mask_sampler: Optional[MaskSampler] = MaskSampler()
mask_roi_aligner: Optional[MaskROIAligner] = MaskROIAligner()
norm_activation: common.NormActivation = common.NormActivation(
norm_momentum=0.997, norm_epsilon=0.0001, use_sync_bn=True)
def test_builder(self, backbone_type, input_size,
segmentation_backbone_type, segmentation_decoder_type,
fusion_type):
num_classes = 2
input_specs = tf.keras.layers.InputSpec(
shape=[None, input_size[0], input_size[1], 3])
segmentation_output_stride = 16
level = int(np.math.log2(segmentation_output_stride))
segmentation_model = semantic_segmentation.SemanticSegmentationModel(
num_classes=2,
backbone=backbones.Backbone(type=segmentation_backbone_type),
decoder=decoders.Decoder(type=segmentation_decoder_type),
head=semantic_segmentation.SegmentationHead(level=level))
model_config = panoptic_maskrcnn_cfg.PanopticMaskRCNN(
num_classes=num_classes,
segmentation_model=segmentation_model,
backbone=backbones.Backbone(type=backbone_type),
shared_backbone=segmentation_backbone_type is None,
shared_decoder=segmentation_decoder_type is None)
l2_regularizer = tf.keras.regularizers.l2(5e-5)
_ = factory.build_panoptic_maskrcnn(input_specs=input_specs,
model_config=model_config,
l2_regularizer=l2_regularizer)
def retinanet_spinenet_mobile_coco() -> cfg.ExperimentConfig:
"""COCO object detection with RetinaNet using Mobile SpineNet backbone."""
train_batch_size = 256
eval_batch_size = 8
steps_per_epoch = COCO_TRAIN_EXAMPLES // train_batch_size
input_size = 384
config = cfg.ExperimentConfig(
runtime=cfg.RuntimeConfig(mixed_precision_dtype='float32'),
task=RetinaNetTask(
annotation_file=os.path.join(COCO_INPUT_PATH_BASE,
'instances_val2017.json'),
model=RetinaNet(
backbone=backbones.Backbone(
type='spinenet_mobile',
spinenet_mobile=backbones.SpineNetMobile(
model_id='49',
stochastic_depth_drop_rate=0.2,
min_level=3,
max_level=7)),
decoder=decoders.Decoder(type='identity',
identity=decoders.Identity()),
head=RetinaNetHead(num_filters=48, use_separable_conv=True),
anchor=Anchor(anchor_size=3),
norm_activation=common.NormActivation(use_sync_bn=True,
activation='swish'),
num_classes=91,
input_size=[input_size, input_size, 3],
min_level=3,
max_level=7),
losses=Losses(l2_weight_decay=3e-5),
train_data=DataConfig(input_path=os.path.join(
COCO_INPUT_PATH_BASE, 'train*'),
is_training=True,
global_batch_size=train_batch_size,
parser=Parser(aug_rand_hflip=True,
aug_scale_min=0.1,
aug_scale_max=2.0)),
validation_data=DataConfig(input_path=os.path.join(
COCO_INPUT_PATH_BASE, 'val*'),
is_training=False,
global_batch_size=eval_batch_size)),
trainer=cfg.TrainerConfig(
train_steps=600 * steps_per_epoch,
validation_steps=COCO_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'stepwise',
'stepwise': {
'boundaries':
[575 * steps_per_epoch, 590 * steps_per_epoch],
'values': [
0.32 * train_batch_size / 256.0,
0.032 * train_batch_size / 256.0,
0.0032 * train_batch_size / 256.0
],
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 2000,
'warmup_learning_rate': 0.0067
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None',
'task.model.min_level == task.model.backbone.spinenet_mobile.min_level',
'task.model.max_level == task.model.backbone.spinenet_mobile.max_level',
])
return config
def seg_deeplabv2_pascal() -> cfg.ExperimentConfig:
"""Image segmentation on imagenet with vggnet & resnet deeplabv2."""
train_batch_size = 16
eval_batch_size = 8
steps_per_epoch = PASCAL_TRAIN_EXAMPLES // train_batch_size
# for Large FOV
fov_dilation_rates = 12
kernel_size = 3
# for ASPP
aspp_dilation_rates = [6, 12, 18, 24]
output_stride = 16
level = int(np.math.log2(output_stride))
config = cfg.ExperimentConfig(
task=SemanticSegmentationTask(
model=SemanticSegmentationModel(
num_classes=21,
input_size=[None, None, 3],
backbone=backbones.Backbone(
type='dilated_vggnet',
dilated_vggnet=backbones.DilatedVGGNet(model_id=16)),
decoder=decoders.Decoder(
type='aspp',
aspp=decoders.ASPP(level=level,
dilation_rates=aspp_dilation_rates,
stem_type='v2',
num_filters=1024,
use_sync_bn=True)),
head=SegmentationHead(level=level,
num_convs=0,
low_level_num_filters=1024,
feature_fusion='deeplabv2'),
norm_activation=common.NormActivation(activation='swish',
norm_momentum=0.9997,
norm_epsilon=1e-3,
use_sync_bn=True)),
losses=Losses(l2_weight_decay=1e-4),
train_data=DataConfig(
input_path=os.path.join(PASCAL_INPUT_PATH_BASE, 'train_aug*'),
# TODO(arashwan): test changing size to 513 to match deeplab.
output_size=[512, 512],
is_training=True,
global_batch_size=train_batch_size,
aug_scale_min=0.5,
aug_scale_max=1.5),
validation_data=DataConfig(input_path=os.path.join(
PASCAL_INPUT_PATH_BASE, 'val*'),
output_size=[512, 512],
is_training=False,
global_batch_size=eval_batch_size,
resize_eval_groundtruth=False,
groundtruth_padded_size=[512, 512],
drop_remainder=False),
# resnet101
init_checkpoint='/home/gunho1123/ckpt_vggnet16_deeplab/',
init_checkpoint_modules='backbone'),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=45 * steps_per_epoch,
validation_steps=PASCAL_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.007,
'decay_steps': 45 * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def seg_deeplabv3plus_ade20k_32(backbone: str,
init_backbone: bool = True
) -> cfg.ExperimentConfig:
"""Semantic segmentation on ADE20K dataset with deeplabv3+."""
epochs = 200
train_batch_size = 128
eval_batch_size = 32
image_size = 512
steps_per_epoch = ADE20K_TRAIN_EXAMPLES // train_batch_size
aspp_dilation_rates = [5, 10, 15]
pretrained_checkpoint_path = BACKBONE_PRETRAINED_CHECKPOINT[
backbone] if init_backbone else None
config = cfg.ExperimentConfig(
task=CustomSemanticSegmentationTaskConfig(
model=base_cfg.SemanticSegmentationModel(
# ADE20K uses only 32 semantic classes for train/evaluation.
# The void (background) class is ignored in train and evaluation.
num_classes=32,
input_size=[None, None, 3],
backbone=Backbone(
type='mobilenet_edgetpu',
mobilenet_edgetpu=MobileNetEdgeTPU(
model_id=backbone,
pretrained_checkpoint_path=pretrained_checkpoint_path,
freeze_large_filters=500,
)),
decoder=decoders.Decoder(
type='aspp',
aspp=decoders.ASPP(
level=BACKBONE_HEADPOINT[backbone],
use_depthwise_convolution=True,
dilation_rates=aspp_dilation_rates,
pool_kernel_size=[256, 256],
num_filters=128,
dropout_rate=0.3,
)),
head=base_cfg.SegmentationHead(
level=BACKBONE_HEADPOINT[backbone],
num_convs=2,
num_filters=256,
use_depthwise_convolution=True,
feature_fusion='deeplabv3plus',
low_level=BACKBONE_LOWER_FEATURES[backbone],
low_level_num_filters=48),
norm_activation=common.NormActivation(activation='relu',
norm_momentum=0.99,
norm_epsilon=2e-3,
use_sync_bn=False)),
train_data=base_cfg.DataConfig(
input_path=os.path.join(ADE20K_INPUT_PATH_BASE, 'train-*'),
output_size=[image_size, image_size],
is_training=True,
global_batch_size=train_batch_size),
validation_data=base_cfg.DataConfig(
input_path=os.path.join(ADE20K_INPUT_PATH_BASE, 'val-*'),
output_size=[image_size, image_size],
is_training=False,
global_batch_size=eval_batch_size,
resize_eval_groundtruth=True,
drop_remainder=False),
evaluation=base_cfg.Evaluation(report_train_mean_iou=False),
),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=epochs * steps_per_epoch,
validation_steps=ADE20K_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'adam',
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.0001,
'decay_steps': epochs * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 4 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def seg_resnetfpn_pascal() -> cfg.ExperimentConfig:
"""Image segmentation on imagenet with resnet-fpn."""
train_batch_size = 256
eval_batch_size = 32
steps_per_epoch = PASCAL_TRAIN_EXAMPLES // train_batch_size
config = cfg.ExperimentConfig(
task=SemanticSegmentationTask(
model=SemanticSegmentationModel(
num_classes=21,
input_size=[512, 512, 3],
min_level=3,
max_level=7,
backbone=backbones.Backbone(
type='resnet', resnet=backbones.ResNet(model_id=50)),
decoder=decoders.Decoder(type='fpn', fpn=decoders.FPN()),
head=SegmentationHead(level=3, num_convs=3),
norm_activation=common.NormActivation(activation='swish',
use_sync_bn=True)),
losses=Losses(l2_weight_decay=1e-4),
train_data=DataConfig(input_path=os.path.join(
PASCAL_INPUT_PATH_BASE, 'train_aug*'),
is_training=True,
global_batch_size=train_batch_size,
aug_scale_min=0.2,
aug_scale_max=1.5),
validation_data=DataConfig(input_path=os.path.join(
PASCAL_INPUT_PATH_BASE, 'val*'),
is_training=False,
global_batch_size=eval_batch_size,
resize_eval_groundtruth=False,
groundtruth_padded_size=[512, 512],
drop_remainder=False),
),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=450 * steps_per_epoch,
validation_steps=PASCAL_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.007,
'decay_steps': 450 * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def seg_deeplabv3plus_cityscapes() -> cfg.ExperimentConfig:
"""Image segmentation on imagenet with resnet deeplabv3+."""
train_batch_size = 16
eval_batch_size = 16
steps_per_epoch = CITYSCAPES_TRAIN_EXAMPLES // train_batch_size
output_stride = 16
aspp_dilation_rates = [6, 12, 18]
multigrid = [1, 2, 4]
stem_type = 'v1'
level = int(np.math.log2(output_stride))
config = cfg.ExperimentConfig(
task=SemanticSegmentationTask(
model=SemanticSegmentationModel(
# Cityscapes uses only 19 semantic classes for train/evaluation.
# The void (background) class is ignored in train and evaluation.
num_classes=19,
input_size=[None, None, 3],
backbone=backbones.Backbone(
type='dilated_resnet',
dilated_resnet=backbones.DilatedResNet(
model_id=101,
output_stride=output_stride,
stem_type=stem_type,
multigrid=multigrid)),
decoder=decoders.Decoder(
type='aspp',
aspp=decoders.ASPP(level=level,
dilation_rates=aspp_dilation_rates,
pool_kernel_size=[512, 1024])),
head=SegmentationHead(level=level,
num_convs=2,
feature_fusion='deeplabv3plus',
low_level=2,
low_level_num_filters=48),
norm_activation=common.NormActivation(activation='swish',
norm_momentum=0.99,
norm_epsilon=1e-3,
use_sync_bn=True)),
losses=Losses(l2_weight_decay=1e-4),
train_data=DataConfig(input_path=os.path.join(
CITYSCAPES_INPUT_PATH_BASE, 'train_fine**'),
crop_size=[512, 1024],
output_size=[1024, 2048],
is_training=True,
global_batch_size=train_batch_size,
aug_scale_min=0.5,
aug_scale_max=2.0),
validation_data=DataConfig(input_path=os.path.join(
CITYSCAPES_INPUT_PATH_BASE, 'val_fine*'),
output_size=[1024, 2048],
is_training=False,
global_batch_size=eval_batch_size,
resize_eval_groundtruth=True,
drop_remainder=False),
# resnet101
init_checkpoint=
'gs://cloud-tpu-checkpoints/vision-2.0/deeplab/deeplab_resnet101_imagenet/ckpt-62400',
init_checkpoint_modules='backbone'),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=500 * steps_per_epoch,
validation_steps=CITYSCAPES_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.01,
'decay_steps': 500 * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def seg_deeplabv3plus_scooter() -> cfg.ExperimentConfig:
"""Image segmentation on scooter dataset with resnet deeplabv3+.
Barebones config for testing purpose (modify batch size, initial lr, steps per epoch, train input path, val input path)
"""
scooter_path_glob = 'D:/data/test_data/val**'
steps_per_epoch = 1
output_stride = 16
aspp_dilation_rates = [6, 12, 18]
multigrid = [1, 2, 4]
stem_type = 'v1'
level = int(np.math.log2(output_stride))
config = cfg.ExperimentConfig(
task=SemanticSegmentationTask(
model=SemanticSegmentationModel(
num_classes=19,
input_size=[512, 512, 3], # specifying this speeds up model inference, no change in size
backbone=backbones.Backbone(
type='dilated_resnet', dilated_resnet=backbones.DilatedResNet(
model_id=101, output_stride=output_stride,
stem_type=stem_type, multigrid=multigrid)),
decoder=decoders.Decoder(
type='aspp',
aspp=decoders.ASPP(
level=level, dilation_rates=aspp_dilation_rates)),
head=SegmentationHead(
level=level,
num_convs=2,
feature_fusion='deeplabv3plus',
low_level=2,
low_level_num_filters=48),
norm_activation=common.NormActivation(
activation='swish',
norm_momentum=0.99,
norm_epsilon=1e-3,
use_sync_bn=True)),
losses=Losses(
l2_weight_decay=1e-4,
ignore_label=250),
train_data=DataConfig(
input_path=scooter_path_glob,
output_size=[512, 512],
is_training=True,
global_batch_size=1,
aug_scale_min=0.5,
aug_scale_max=2.0),
validation_data=DataConfig(
input_path=scooter_path_glob,
output_size=[512, 512],
is_training=False,
global_batch_size=1,
resize_eval_groundtruth=True,
drop_remainder=False)),
# resnet101
# init_checkpoint='D:/repos/data_root/test_data/deeplab_cityscapes_pretrained/model.ckpt',
# init_checkpoint_modules='all'),
# init_checkpoint='gs://cloud-tpu-checkpoints/vision-2.0/deeplab/deeplab_resnet101_imagenet/ckpt-62400',
# init_checkpoint_modules='backbone'),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=500 * steps_per_epoch,
validation_steps=1021,
validation_interval=steps_per_epoch,
continuous_eval_timeout=1,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.007,
'decay_steps': 500 * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def multitask_vision() -> multi_cfg.MultiTaskExperimentConfig:
"""
Vision task with single backbone and multiple heads.
Each head can be a segmenter, detector or classifier.
TODO: use same num_class and input_size in both task and model definition
multi_cfg.MultiTaskConfig:
- Retains each task_name, entire task, eval_steps and weights,
- Entire_task used in respective multitask trainers for train_step
- Weights used in task_sampler
multi_cfg.MultiTaskTrainerConfig:
- trainer_type and task_sampler used to configure task sampling in train_lib
- Normal multi_cfg.TrainerConfig params used directly in train_lib
"""
input_path_segmentation = ''
input_path_classification = ''
input_path_yolo = ''
steps_per_epoch = 6915 + 2486 + 600
train_batch_size = 1
eval_batch_size = 1
validation_steps = 1021 + 621 + 600
segmentation_routine = multi_cfg.TaskRoutine(
task_name='segmentation',
task_config=SemanticSegmentationSubtask(
model=SemanticSegmentationModelSpecs(num_classes=19,
input_size=[256, 256, 3]),
losses=SegmentationLosses(ignore_label=250,
top_k_percent_pixels=0.3),
train_data=SegmentationDataConfig(
output_size=[256, 256],
input_path=input_path_segmentation,
global_batch_size=train_batch_size,
is_training=True,
aug_scale_min=0.5,
aug_scale_max=2.0,
preserve_aspect_ratio=False,
aug_policy='randaug',
randaug_magnitude=5,
randaug_available_ops=[
'AutoContrast', 'Equalize', 'Invert', 'Rotate',
'Posterize', 'Solarize', 'Color', 'Contrast', 'Brightness',
'Sharpness', 'Cutout', 'SolarizeAdd'
]),
validation_data=SegmentationDataConfig(
output_size=[256, 256],
input_path=input_path_segmentation,
global_batch_size=eval_batch_size,
is_training=False,
resize_eval_groundtruth=True,
drop_remainder=False)),
eval_steps=603, # check where eval steps is used
task_weight=1.0)
classification_routine = multi_cfg.TaskRoutine(
task_name='classification',
task_config=ImageClassificationSubtask(
model=ImageClassificationModelSpecs(num_classes=4,
input_size=[256, 256, 3]),
losses=ClassificationLosses(label_smoothing=0.1),
train_data=ClassificationDataConfig(
input_path=input_path_classification,
is_training=True,
global_batch_size=train_batch_size,
aug_policy='randaug',
randaug_magnitude=5),
validation_data=ClassificationDataConfig(
input_path=input_path_classification,
is_training=False,
global_batch_size=eval_batch_size,
drop_remainder=False)),
eval_steps=621, # check where eval steps is used
task_weight=1.0)
yolo_routine = multi_cfg.TaskRoutine(
task_name='yolo',
task_config=YoloSubtask(
model=YoloModelSpecs(num_classes=4,
input_size=[256, 256, 3],
head=YoloHead(anchor_per_scale=3,
strides=[16, 32, 64],
anchors=[
12, 16, 19, 36, 40, 28, 36,
75, 76, 55, 72, 146, 142,
110, 192, 243, 459, 401
],
xy_scale=[1.2, 1.1, 1.05])),
losses=YoloLosses(l2_weight_decay=1e-4, iou_loss_thres=0.5),
train_data=YoloDataConfig(input_path=input_path_yolo,
is_training=True,
global_batch_size=train_batch_size,
aug_policy='randaug',
randaug_magnitude=5),
validation_data=YoloDataConfig(input_path=input_path_yolo,
is_training=False,
global_batch_size=eval_batch_size,
drop_remainder=False)),
eval_steps=600, # check where eval steps is used
task_weight=1.0)
model_config = MultiHeadModel(
input_size=[256, 256, 3],
backbone=backbones.Backbone(type='hardnet',
hardnet=backbones.HardNet(model_id=70)),
norm_activation=common.NormActivation(activation='relu',
norm_momentum=0.9997,
norm_epsilon=0.001,
use_sync_bn=True),
heads=[
Submodel(
name='classification',
num_classes=4,
head=ImageClassificationHead(
level=0, # decoder is identity function
num_convs=2,
num_filters=256,
add_head_batch_norm=False,
dropout_rate=0.2)),
Submodel(name='segmentation',
num_classes=19,
decoder=decoders.Decoder(
type='hardnet',
hardnet=decoders.HardNet(model_id=70)),
head=SegmentationHead(level=0,
num_convs=0,
feature_fusion=None,
low_level=0,
low_level_num_filters=0)),
Submodel(name='yolo',
num_classes=4,
decoder=decoders.Decoder(type='pan',
pan=decoders.PAN(levels=3)),
head=YoloHead(anchor_per_scale=3,
strides=[16, 32, 64],
anchors=[
12, 16, 19, 36, 40, 28, 36, 75, 76, 55,
72, 146, 142, 110, 192, 243, 459, 401
],
xy_scale=[1.2, 1.1, 1.05]))
],
l2_weight_decay=1e-4)
return multi_cfg.MultiTaskExperimentConfig(
task=multi_cfg.MultiTaskConfig(model=model_config,
init_checkpoint=None,
task_routines=(segmentation_routine,
classification_routine,
yolo_routine)),
trainer=multi_cfg.MultiTaskTrainerConfig(
trainer_type="interleaving",
task_sampler=multi_cfg.TaskSamplingConfig(
type="proportional",
proportional=multi_cfg.ProportionalSampleConfig(
alpha=1.0)), # uniform, proportional or annealing
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=45 * steps_per_epoch,
validation_steps=validation_steps,
validation_interval=steps_per_epoch,
best_checkpoint_eval_metric='mean_iou',
continuous_eval_timeout=3600,
max_to_keep=5,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.007,
'decay_steps': 45 * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})))
def maskrcnn_spinenet_coco() -> cfg.ExperimentConfig:
"""COCO object detection with Mask R-CNN with SpineNet backbone."""
steps_per_epoch = 463
coco_val_samples = 5000
config = cfg.ExperimentConfig(
runtime=cfg.RuntimeConfig(mixed_precision_dtype='bfloat16'),
task=MaskRCNNTask(
annotation_file=os.path.join(COCO_INPUT_PATH_BASE,
'instances_val2017.json'),
model=MaskRCNN(
backbone=backbones.Backbone(
type='spinenet',
spinenet=backbones.SpineNet(model_id='49')),
decoder=decoders.Decoder(type='identity',
identity=decoders.Identity()),
anchor=Anchor(anchor_size=3),
norm_activation=common.NormActivation(use_sync_bn=True),
num_classes=91,
input_size=[640, 640, 3],
min_level=3,
max_level=7,
include_mask=True),
losses=Losses(l2_weight_decay=0.00004),
train_data=DataConfig(input_path=os.path.join(
COCO_INPUT_PATH_BASE, 'train*'),
is_training=True,
global_batch_size=256,
parser=Parser(aug_rand_hflip=True,
aug_scale_min=0.5,
aug_scale_max=2.0)),
validation_data=DataConfig(input_path=os.path.join(
COCO_INPUT_PATH_BASE, 'val*'),
is_training=False,
global_batch_size=8)),
trainer=cfg.TrainerConfig(
train_steps=steps_per_epoch * 350,
validation_steps=coco_val_samples // 8,
validation_interval=steps_per_epoch,
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'stepwise',
'stepwise': {
'boundaries':
[steps_per_epoch * 320, steps_per_epoch * 340],
'values': [0.28, 0.028, 0.0028],
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 2000,
'warmup_learning_rate': 0.0067
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def basnet_duts() -> cfg.ExperimentConfig:
"""Image segmentation on imagenet with resnet deeplabv3."""
train_batch_size = 16
eval_batch_size = 16
steps_per_epoch = DUTS_TRAIN_EXAMPLES // train_batch_size
config = cfg.ExperimentConfig(
task=BASNetTask(
model=BASNetModel(
#num_classes=21,
# TODO(arashwan): test changing size to 513 to match deeplab.
input_size=[224, 224, 3], # Resize to 256, 256
backbone=backbones.Backbone(
type='basnet_en', basnet_en=backbones.BASNet_En(
)),
decoder=decoders.Decoder(
type='basnet_de', basnet_de=decoders.BASNet_De(
)),
#head=BASNetHead(level=3, num_convs=0),
norm_activation=common.NormActivation(
activation='relu',
norm_momentum=0.99,
norm_epsilon=1e-3,
use_sync_bn=True)),
losses=Losses(l2_weight_decay=0),
train_data=DataConfig(
#input_path=os.path.join(PASCAL_INPUT_PATH_BASE, 'train_aug*'), # Dataset Path #
input_path=os.path.join(DUTS_INPUT_PATH_BASE_TR, 'DUTS-TR-*'),
is_training=True,
global_batch_size=train_batch_size,
#aug_scale_min=0.5,
#aug_scale_max=2.0
),
validation_data=DataConfig(
input_path=os.path.join(DUTS_INPUT_PATH_BASE_VAL, 'DUTS-TE-*'),
is_training=False,
global_batch_size=eval_batch_size,
),
#init_checkpoint='',
#init_checkpoint_modules='backbone'
),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=500 * steps_per_epoch, # (gunho) more epochs
validation_steps=DUTS_VAL_EXAMPLES // eval_batch_size, # No validation in BASNet
validation_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'adam', #BASNet
'adam': {
'beta_1': 0.9,
'beta_2': 0.999,
'epsilon': 1e-8,
}
},
'learning_rate': {
'type': 'stepwise',
'stepwise': {'boundaries': [70*steps_per_epoch, 100*steps_per_epoch, 150*steps_per_epoch],
'values': [0.01, 0.001, 0.001, 0.0001]}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def mnv2_deeplabv3_cityscapes() -> cfg.ExperimentConfig:
"""Image segmentation on cityscapes with mobilenetv2 deeplabv3."""
train_batch_size = 16
eval_batch_size = 16
steps_per_epoch = CITYSCAPES_TRAIN_EXAMPLES // train_batch_size
output_stride = 16
aspp_dilation_rates = []
pool_kernel_size = [512, 1024]
level = int(np.math.log2(output_stride))
config = cfg.ExperimentConfig(
task=SemanticSegmentationTask(
model=SemanticSegmentationModel(
# Cityscapes uses only 19 semantic classes for train/evaluation.
# The void (background) class is ignored in train and evaluation.
num_classes=19,
input_size=[None, None, 3],
backbone=backbones.Backbone(type='mobilenet',
mobilenet=backbones.MobileNet(
model_id='MobileNetV2',
output_stride=output_stride)),
decoder=decoders.Decoder(
type='aspp',
aspp=decoders.ASPP(level=level,
dilation_rates=aspp_dilation_rates,
pool_kernel_size=pool_kernel_size)),
head=SegmentationHead(level=level, num_convs=0),
norm_activation=common.NormActivation(activation='relu',
norm_momentum=0.99,
norm_epsilon=1e-3,
use_sync_bn=True)),
losses=Losses(l2_weight_decay=4e-5),
train_data=DataConfig(input_path=os.path.join(
CITYSCAPES_INPUT_PATH_BASE, 'train_fine**'),
crop_size=[512, 1024],
output_size=[1024, 2048],
is_training=True,
global_batch_size=train_batch_size,
aug_scale_min=0.5,
aug_scale_max=2.0),
validation_data=DataConfig(input_path=os.path.join(
CITYSCAPES_INPUT_PATH_BASE, 'val_fine*'),
output_size=[1024, 2048],
is_training=False,
global_batch_size=eval_batch_size,
resize_eval_groundtruth=True,
drop_remainder=False),
# Coco pre-trained mobilenetv2 checkpoint
init_checkpoint=
'gs://tf_model_garden/cloud/vision-2.0/deeplab/deeplabv3_mobilenetv2_coco/best_ckpt-63',
init_checkpoint_modules='backbone'),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=100000,
validation_steps=CITYSCAPES_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
best_checkpoint_eval_metric='mean_iou',
best_checkpoint_export_subdir='best_ckpt',
best_checkpoint_metric_comp='higher',
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.01,
'decay_steps': 100000,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def cascadercnn_spinenet_coco() -> cfg.ExperimentConfig:
"""COCO object detection with Cascade RCNN-RS with SpineNet backbone."""
steps_per_epoch = 463
coco_val_samples = 5000
train_batch_size = 256
eval_batch_size = 8
config = cfg.ExperimentConfig(
runtime=cfg.RuntimeConfig(mixed_precision_dtype='bfloat16'),
task=MaskRCNNTask(
annotation_file=os.path.join(COCO_INPUT_PATH_BASE,
'instances_val2017.json'),
model=MaskRCNN(
backbone=backbones.Backbone(
type='spinenet',
spinenet=backbones.SpineNet(
model_id='49',
min_level=3,
max_level=7,
)),
decoder=decoders.Decoder(
type='identity', identity=decoders.Identity()),
roi_sampler=ROISampler(cascade_iou_thresholds=[0.6, 0.7]),
detection_head=DetectionHead(
class_agnostic_bbox_pred=True, cascade_class_ensemble=True),
anchor=Anchor(anchor_size=3),
norm_activation=common.NormActivation(
use_sync_bn=True, activation='swish'),
num_classes=91,
input_size=[640, 640, 3],
min_level=3,
max_level=7,
include_mask=True),
losses=Losses(l2_weight_decay=0.00004),
train_data=DataConfig(
input_path=os.path.join(COCO_INPUT_PATH_BASE, 'train*'),
is_training=True,
global_batch_size=train_batch_size,
parser=Parser(
aug_rand_hflip=True, aug_scale_min=0.1, aug_scale_max=2.5)),
validation_data=DataConfig(
input_path=os.path.join(COCO_INPUT_PATH_BASE, 'val*'),
is_training=False,
global_batch_size=eval_batch_size,
drop_remainder=False)),
trainer=cfg.TrainerConfig(
train_steps=steps_per_epoch * 500,
validation_steps=coco_val_samples // eval_batch_size,
validation_interval=steps_per_epoch,
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'stepwise',
'stepwise': {
'boundaries': [
steps_per_epoch * 475, steps_per_epoch * 490
],
'values': [0.32, 0.032, 0.0032],
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 2000,
'warmup_learning_rate': 0.0067
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None',
'task.model.min_level == task.model.backbone.spinenet.min_level',
'task.model.max_level == task.model.backbone.spinenet.max_level',
])
return config
def seg_deeplabv3_pascal() -> cfg.ExperimentConfig:
"""Image segmentation on imagenet with resnet deeplabv3."""
train_batch_size = 16
eval_batch_size = 8
steps_per_epoch = PASCAL_TRAIN_EXAMPLES // train_batch_size
output_stride = 16
aspp_dilation_rates = [12, 24, 36] # [6, 12, 18] if output_stride = 16
multigrid = [1, 2, 4]
stem_type = 'v1'
level = int(np.math.log2(output_stride))
config = cfg.ExperimentConfig(
task=SemanticSegmentationTask(
model=SemanticSegmentationModel(
num_classes=21,
input_size=[None, None, 3],
backbone=backbones.Backbone(
type='dilated_resnet',
dilated_resnet=backbones.DilatedResNet(
model_id=101,
output_stride=output_stride,
multigrid=multigrid,
stem_type=stem_type)),
decoder=decoders.Decoder(
type='aspp',
aspp=decoders.ASPP(level=level,
dilation_rates=aspp_dilation_rates)),
head=SegmentationHead(level=level, num_convs=0),
norm_activation=common.NormActivation(activation='swish',
norm_momentum=0.9997,
norm_epsilon=1e-3,
use_sync_bn=True)),
losses=Losses(l2_weight_decay=1e-4),
train_data=DataConfig(
input_path=os.path.join(PASCAL_INPUT_PATH_BASE, 'train_aug*'),
# TODO(arashwan): test changing size to 513 to match deeplab.
output_size=[512, 512],
is_training=True,
global_batch_size=train_batch_size,
aug_scale_min=0.5,
aug_scale_max=2.0),
validation_data=DataConfig(input_path=os.path.join(
PASCAL_INPUT_PATH_BASE, 'val*'),
output_size=[512, 512],
is_training=False,
global_batch_size=eval_batch_size,
resize_eval_groundtruth=False,
groundtruth_padded_size=[512, 512],
drop_remainder=False),
# resnet101
init_checkpoint=
'gs://cloud-tpu-checkpoints/vision-2.0/deeplab/deeplab_resnet101_imagenet/ckpt-62400',
init_checkpoint_modules='backbone'),
trainer=cfg.TrainerConfig(
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
train_steps=45 * steps_per_epoch,
validation_steps=PASCAL_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.007,
'decay_steps': 45 * steps_per_epoch,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 5 * steps_per_epoch,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def detector_yolo() -> cfg.ExperimentConfig:
"""YOLO on custom datasets"""
config = cfg.ExperimentConfig(
task=YoloTask(
model=YoloModel(
num_classes=6,
input_size=[256, 256, 3],
backbone=backbones.Backbone(
type='hardnet', hardnet=backbones.HardNet(model_id=70)),
decoder=decoders.Decoder(
type='pan', pan=decoders.PAN(levels=3)),
head=YoloHead(
anchor_per_scale=3,
strides=[16, 32, 64],
anchors=[12,16, 19,36, 40,28, 36,75, 76,55, 72,146, 142,110, 192,243, 459,401],
xy_scale=[1.2, 1.1, 1.05]
),
norm_activation=common.NormActivation(
activation='relu',
norm_momentum=0.9997,
norm_epsilon=0.001,
use_sync_bn=True)),
losses=YoloLosses(l2_weight_decay=1e-4,
iou_loss_thres=0.5),
train_data=DataConfig(
input_path='D:/data/whizz_tf/detect_env*',
output_size=[256, 256],
is_training=True,
global_batch_size=1),
validation_data=DataConfig(
input_path='D:/data/whizz_tf/detect_env*',
output_size=[256, 256],
is_training=False,
global_batch_size=1,
drop_remainder=False),
# init_checkpoint=None
init_checkpoint_modules='backbone'),
trainer=cfg.TrainerConfig(
steps_per_loop=2,
summary_interval=2,
checkpoint_interval=2,
train_steps=20,
validation_steps=20,
validation_interval=2,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'polynomial',
'polynomial': {
'initial_learning_rate': 0.007,
'decay_steps': 20,
'end_learning_rate': 0.0,
'power': 0.9
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 2,
'warmup_learning_rate': 0
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
def retinanet_spinenet_coco() -> cfg.ExperimentConfig:
"""COCO object detection with RetinaNet using SpineNet backbone."""
train_batch_size = 256
eval_batch_size = 8
steps_per_epoch = COCO_TRIAN_EXAMPLES // train_batch_size
input_size = 640
config = cfg.ExperimentConfig(
runtime=cfg.RuntimeConfig(mixed_precision_dtype='float32'),
task=RetinaNetTask(model=RetinaNet(
backbone=backbones.Backbone(
type='spinenet', spinenet=backbones.SpineNet(model_id='49')),
decoder=decoders.Decoder(type='identity',
identity=decoders.Identity()),
anchor=Anchor(anchor_size=3),
norm_activation=common.NormActivation(use_sync_bn=True),
num_classes=91,
input_size=[input_size, input_size, 3],
min_level=3,
max_level=7),
losses=Losses(l2_weight_decay=4e-5),
train_data=DataConfig(
input_path=os.path.join(COCO_INPUT_PATH_BASE,
'train*'),
is_training=True,
global_batch_size=train_batch_size,
parser=Parser(aug_rand_hflip=True,
aug_scale_min=0.5,
aug_scale_max=2.0)),
validation_data=DataConfig(
input_path=os.path.join(COCO_INPUT_PATH_BASE,
'val*'),
is_training=False,
global_batch_size=eval_batch_size)),
trainer=cfg.TrainerConfig(
train_steps=350 * steps_per_epoch,
validation_steps=COCO_VAL_EXAMPLES // eval_batch_size,
validation_interval=steps_per_epoch,
steps_per_loop=steps_per_epoch,
summary_interval=steps_per_epoch,
checkpoint_interval=steps_per_epoch,
optimizer_config=optimization.OptimizationConfig({
'optimizer': {
'type': 'sgd',
'sgd': {
'momentum': 0.9
}
},
'learning_rate': {
'type': 'stepwise',
'stepwise': {
'boundaries':
[320 * steps_per_epoch, 340 * steps_per_epoch],
'values': [
0.28 * train_batch_size / 256.0,
0.028 * train_batch_size / 256.0,
0.0028 * train_batch_size / 256.0
],
}
},
'warmup': {
'type': 'linear',
'linear': {
'warmup_steps': 2000,
'warmup_learning_rate': 0.0067
}
}
})),
restrictions=[
'task.train_data.is_training != None',
'task.validation_data.is_training != None'
])
return config
