def __init__(self):
'''
Performs the photometric distortions defined by the `train_transform_param` instructions
of the original Caffe implementation of SSD.
'''
self.convert_RGB_to_HSV = ConvertColor(current='RGB', to='HSV')
self.convert_HSV_to_RGB = ConvertColor(current='HSV', to='RGB')
self.convert_to_float32 = ConvertDataType(to='float32')
self.convert_to_uint8 = ConvertDataType(to='uint8')
self.convert_to_3_channels = ConvertTo3Channels()
self.random_brightness = RandomBrightness(lower=-32,
upper=32,
prob=0.5)
self.random_contrast = RandomContrast(lower=0.5, upper=1.5, prob=0.5)
self.random_saturation = RandomSaturation(lower=0.5,
upper=1.5,
prob=0.5)
self.random_hue = RandomHue(max_delta=18, prob=0.5)
self.random_channel_swap = RandomChannelSwap(prob=0.0)
self.sequence1 = [
self.convert_to_3_channels, self.convert_to_float32,
self.random_brightness, self.random_contrast,
self.convert_to_uint8, self.convert_RGB_to_HSV,
self.convert_to_float32, self.random_saturation, self.random_hue,
self.convert_to_uint8, self.convert_HSV_to_RGB,
self.random_channel_swap
]
self.sequence2 = [
self.convert_to_3_channels, self.convert_to_float32,
self.random_brightness, self.convert_to_uint8,
self.convert_RGB_to_HSV, self.convert_to_float32,
self.random_saturation, self.random_hue, self.convert_to_uint8,
self.convert_HSV_to_RGB, self.convert_to_float32,
self.random_contrast, self.convert_to_uint8,
self.random_channel_swap
]
def __init__(self):
# Variables to hold the description of the experiment
self.config_description = "This is the template config file."
# System dependent variable
self._workers = 5
self._multiprocessing = True
# Variables for comet.ml
self._project_name = "jpeg_deep"
self._workspace = "ssd"
# Network variables
self._weights = "/dlocal/home/2017018/bdegue01/weights/jpeg_deep/classification_dct/lcrfa_y/classification_dct_jpeg-deep_46lWNDTTTHMqAvNj88wqA2HPofUMhHva/checkpoints/epoch-71_loss-0.9617_val_loss-1.6794.h5"
self._network = SSD300_resnet(
n_classes=80,
scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05],
backbone="lcrfa_y")
# Training variables
self._epochs = 240
self._batch_size = 32
self._steps_per_epoch = 3700
self._validation_steps = 156
self.optimizer_parameters = {"lr": 0.001, "momentum": 0.9}
self._optimizer = SGD(**self.optimizer_parameters)
self._loss = SSDLoss(neg_pos_ratio=3, alpha=1.0).compute_loss
self._metrics = None
dataset_path = environ["DATASET_PATH"]
self.train_image_dir = join(dataset_path, "train2017")
self.train_annotation_path = join(
dataset_path, "annotations/instances_train2017.json")
self.validation_image_dir = join(dataset_path, "val2017")
self.validation_annotation_path = join(
dataset_path, "annotations/instances_val2017.json")
self.test_annotation_path = join(
dataset_path, "annotations/image_info_test-dev2017.json")
self.test_image_dir = join(dataset_path, "test2017")
# Keras stuff
self.model_checkpoint = None
self.reduce_lr_on_plateau = ReduceLROnPlateau(patience=5, verbose=1)
self.terminate_on_nan = TerminateOnNaN()
self.early_stopping = EarlyStopping(monitor='val_loss',
min_delta=0,
patience=15)
self._callbacks = [
self.reduce_lr_on_plateau, self.early_stopping,
self.terminate_on_nan
]
self.input_encoder = SSDInputEncoder(
n_classes=80, scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05])
self.train_transformations = [SSDDataAugmentation()]
self.validation_transformations = [
ConvertTo3Channels(),
Resize(height=300, width=300)
]
self.test_transformations = [
ConvertTo3Channels(),
Resize(height=300, width=300)
]
self._train_generator = None
self._validation_generator = None
self._test_generator = None
self._horovod = None
self.coco_classes = [
"background", "person", "bicycle", "car", "motorcycle", "airplane",
"bus", "train", "truck", "boat", "traffic light", "fire hydrant",
"stop sign", "parking meter", "bench", "bird", "cat", "dog",
"horse", "sheep", "cow", "elephant", "bear", "zebra", "giraffe",
"backpack", "umbrella", "handbag", "tie", "suitcase", "frisbee",
"skis", "snowboard", "sports ball", "kite", "baseball bat",
"baseball glove", "skateboard", "surfboard", "tennis racket",
"bottle", "wine glass", "cup", "fork", "knife", "spoon", "bowl",
"banana", "apple", "sandwich", "orange", "broccoli", "carrot",
"hot dog", "pizza", "donut", "cake", "chair", "couch",
"potted plant", "bed", "dining table", "toilet", "tv", "laptop",
"mouse", "remote", "keyboard", "cell phone", "microwave", "oven",
"toaster", "sink", "refrigerator", "book", "clock", "vase",
"scissors", "teddy bear", "hair drier", "toothbrush"
]
self._displayer = DisplayerObjects(classes=self.coco_classes)
def __init__(self):
# Variables to hold the description of the experiment
self.config_description = "This is the template config file."
# System dependent variable
self._workers = 5
self._multiprocessing = True
# Variables for comet.ml
self._project_name = "jpeg_deep"
self._workspace = "ssd"
# Network variables
self._weights = "/dlocal/home/2017018/bdegue01/weights/jpeg_deep/classification_dct/lcrfa_y/classification_dct_jpeg-deep_46lWNDTTTHMqAvNj88wqA2HPofUMhHva/checkpoints/epoch-71_loss-0.9617_val_loss-1.6794.h5"
self._network = SSD300_resnet(backbone="lcrfa_y",
dct=True,
image_shape=(38, 38))
# Training variables
self._epochs = 240
self._batch_size = 32
self._steps_per_epoch = 1000
self.optimizer_parameters = {"lr": 0.001, "momentum": 0.9}
self._optimizer = SGD(**self.optimizer_parameters)
self._loss = SSDLoss(neg_pos_ratio=3, alpha=1.0).compute_loss
self._metrics = None
dataset_path = environ["DATASET_PATH"]
images_2007_path = join(dataset_path, "VOC2007/JPEGImages")
images_2012_path = join(dataset_path, "VOC2012/JPEGImages")
self.train_sets = [(images_2007_path,
join(dataset_path,
"VOC2007/ImageSets/Main/train.txt")),
(images_2012_path,
join(dataset_path,
"VOC2012/ImageSets/Main/train.txt"))]
self.validation_sets = [(images_2007_path,
join(dataset_path,
"VOC2007/ImageSets/Main/val.txt")),
(images_2012_path,
join(dataset_path,
"VOC2012/ImageSets/Main/val.txt"))]
self.test_sets = [(images_2007_path,
join(dataset_path,
"VOC2007/ImageSets/Main/test.txt"))]
# Keras stuff
self.model_checkpoint = None
self.reduce_lr_on_plateau = ReduceLROnPlateau(patience=5, verbose=1)
self.terminate_on_nan = TerminateOnNaN()
self.early_stopping = EarlyStopping(monitor='val_loss',
min_delta=0,
patience=15)
self._callbacks = [
self.reduce_lr_on_plateau, self.early_stopping,
self.terminate_on_nan
]
self.input_encoder = SSDInputEncoder()
self.train_transformations = [SSDDataAugmentation()]
self.validation_transformations = [
ConvertTo3Channels(),
Resize(height=300, width=300)
]
self.test_transformations = [
ConvertTo3Channels(),
Resize(height=300, width=300)
]
self._train_generator = None
self._validation_generator = None
self._test_generator = None
self._horovod = None
self._displayer = DisplayerObjects()
def __init__(self):
# Variables to hold the description of the experiment
self.config_description = "This is the template config file."
# System dependent variable
self._workers = 5
self._multiprocessing = True
# Variables for comet.ml
self._project_name = "jpeg_deep"
self._workspace = "ssd"
# Network variables
self._weights = "/dlocal/home/2017018/bdegue01/weights/jpeg_deep/classification_dct/lcrfa_y/classification_dct_jpeg-deep_46lWNDTTTHMqAvNj88wqA2HPofUMhHva/checkpoints/epoch-71_loss-0.9617_val_loss-1.6794.h5"
self._network = SSD300_resnet(
n_classes=80,
scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05],
backbone="lcrfa_y")
# Training variables
self._epochs = 240
self._batch_size = 32
self._steps_per_epoch = 3700
self._validation_steps = 156
self.optimizer_parameters = {"lr": 0.001, "momentum": 0.9}
self._optimizer = SGD(**self.optimizer_parameters)
self._loss = SSDLoss(neg_pos_ratio=3, alpha=1.0).compute_loss
self._metrics = None
dataset_path = environ["DATASET_PATH"]
self.train_image_dir = join(dataset_path, "train2017")
self.train_annotation_path = join(
dataset_path, "annotations/instances_train2017.json")
self.validation_image_dir = join(dataset_path, "val2017")
self.validation_annotation_path = join(
dataset_path, "annotations/instances_val2017.json")
# Keras stuff
self.model_checkpoint = None
self.reduce_lr_on_plateau = ReduceLROnPlateau(patience=5, verbose=1)
self.terminate_on_nan = TerminateOnNaN()
self.early_stopping = EarlyStopping(monitor='val_loss',
min_delta=0,
patience=15)
self._callbacks = [
self.reduce_lr_on_plateau, self.early_stopping,
self.terminate_on_nan
]
self.input_encoder = SSDInputEncoder(
n_classes=80, scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05])
self.train_tranformations = [SSDDataAugmentation()]
self.validation_transformations = [
ConvertTo3Channels(),
Resize(height=300, width=300)
]
self.test_transformations = [
ConvertTo3Channels(),
Resize(height=300, width=300)
]
self._train_generator = None
self._validation_generator = None
self._test_generator = None
self._horovod = None
def __init__(self):
# Variables to hold the description of the experiment
self.config_description = "Configuration file for the training on the PascalVOC 07 dataset."
# System dependent variable
self._workers = 5
self._multiprocessing = True
# Variables for comet.ml
self._project_name = "jpeg_deep"
self._workspace = "ssd"
# Network variables
self._weights = "/dlocal/home/2017018/bdegue01/weights/jpeg_deep/reproduce/vgg/full_reg/vggd/epoch-86_loss-1.4413_val_loss-1.9857_ssd.h5"
self._network = SSD300()
# Training variables
self._epochs = 240
self._batch_size = 32
self._steps_per_epoch = 1000
self.optimizer_parameters = {
"lr": 0.001, "momentum": 0.9}
self._optimizer = SGD(**self.optimizer_parameters)
self._loss = SSDLoss(neg_pos_ratio=3, alpha=1.0).compute_loss
self._metrics = None
dataset_path = environ["DATASET_PATH"]
images_2007_path = join(dataset_path, "VOC2007/JPEGImages")
self.train_sets = [(images_2007_path, join(
dataset_path, "VOC2007/ImageSets/Main/train.txt"))]
self.validation_sets = [(images_2007_path, join(
dataset_path, "VOC2007/ImageSets/Main/val.txt"))]
self.test_sets = [(images_2007_path, join(
dataset_path, "VOC2007/ImageSets/Main/test.txt"))]
# Keras stuff
self.model_checkpoint = None
self.reduce_lr_on_plateau = ReduceLROnPlateau(patience=5, verbose=1)
self.terminate_on_nan = TerminateOnNaN()
self.early_stopping = EarlyStopping(monitor='val_loss',
min_delta=0,
patience=15)
self._callbacks = [self.reduce_lr_on_plateau, self.early_stopping,
self.terminate_on_nan]
self.input_encoder = SSDInputEncoder()
self.train_transformations = [SSDDataAugmentation()]
self.validation_transformations = [
ConvertTo3Channels(), Resize(height=300, width=300)]
self.test_transformations = [ConvertTo3Channels(), Resize(
height=300, width=300)]
self._train_generator = None
self._validation_generator = None
self._test_generator = None
self._displayer = DisplayerObjects()
