def dataset(self, config, debug=False):
# #### load dataset
train_dataset = coco.CocoDataSet('./COCO2017',
'train',
flip_ratio=0.,
pad_mode='fixed',
config=config,
debug=False)
print(train_dataset.get_categories())
assert config.NUM_CLASSES == len(
train_dataset.get_categories()
), f"NUM_CLASSES must be compare with dataset, set:{config.NUM_CLASSES} != {len(train_dataset.get_categories())}"
train_generator = data_generator.DataGenerator(train_dataset)
train_tf_dataset = tf.data.Dataset.from_generator(
train_generator, (tf.float32, tf.float32, tf.float32, tf.int32,
tf.float32, tf.float32))
self.train_tf_dataset = train_tf_dataset.padded_batch(
config.IMAGES_PER_GPU,
padded_shapes=(
[None, None, None], # img
[None], #img_meta
[None, None], #bboxes
[None], #labels
[None, None, None], #masks
[None, None, 1])) # global_mask
eval_dataset = coco.CocoDataSet('./COCO2017',
'val',
flip_ratio=0.,
pad_mode='fixed',
config=config,
debug=False)
eval_generator = data_generator.DataGenerator(eval_dataset)
eval_tf_dataset = tf.data.Dataset.from_generator(
eval_generator, (tf.float32, tf.float32, tf.float32, tf.int32,
tf.float32, tf.float32))
self.eval_tf_dataset = eval_tf_dataset.padded_batch(
config.IMAGES_PER_GPU,
padded_shapes=(
[None, None, None], # img
[None], #img_meta
[None, None], #bboxes
[None], #labels
[None, None, None], #masks
[None, None, 1])) # global_mask
if debug:
idx = np.random.choice(range(len(train_dataset)))
img, img_meta, bboxes, labels, masks, global_mask = train_dataset[
idx]
rgb_img = np.round(img + config.MEAN_PIXEL)
ori_img = utils.get_original_image(img, img_meta,
config.MEAN_PIXEL)
visualize.display_instances(rgb_img, bboxes, labels,
train_dataset.get_categories())
self.train_dataset = train_dataset
from detection.models.detectors import faster_rcnn
print("Num GPUs Available: ",
len(tf.config.experimental.list_physical_devices('GPU')))
for gpu in tf.config.experimental.list_physical_devices('GPU'):
tf.config.experimental.set_memory_growth(gpu, True)
# load dataset
from detection.datasets import coco, data_generator
img_mean = (123.675, 116.28, 103.53)
# img_std = (58.395, 57.12, 57.375)
img_std = (1., 1., 1.)
train_dataset = coco.CocoDataSet('./COCO2017/',
'val',
flip_ratio=0.5,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(640, 832),
debug=True)
train_generator = data_generator.DataGenerator(train_dataset)
# load model
from detection.models.detectors import faster_rcnn
strategy = tf.distribute.MirroredStrategy()
with strategy.scope():
model = faster_rcnn.FasterRCNN(
num_classes=len(train_dataset.get_categories()))
# # Currently, memory growth needs to be the same across GPUs
# for gpu in gpus:
# tf.config.experimental.set_memory_growth(gpu, True)
# logical_gpus = tf.config.experimental.list_logical_devices('GPU')
# print(len(gpus), "Physical GPUs,", len(logical_gpus), "Logical GPUs")
# except RuntimeError as e:
# # Memory growth must be set before GPUs have been initialized
# print(e)
img_mean = (123.675, 116.28, 103.53)
# img_std = (58.395, 57.12, 57.375)
img_std = (1., 1., 1.)
val_dataset = coco.CocoDataSet('./COCO2017/', 'val',
flip_ratio=0,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(800, 1344))
print(len(val_dataset))
model = faster_rcnn.FasterRCNN(
num_classes=len(val_dataset.get_categories()))
img, img_meta, bboxes, labels = val_dataset[0]
batch_imgs = tf.Variable(np.expand_dims(img, 0))
batch_metas = tf.Variable(np.expand_dims(img_meta, 0))
_ = model((batch_imgs, batch_metas), training=False)
# memory region needed by the TensorFlow process
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
# load dataset
from detection.datasets import coco
img_mean = (123.675, 116.28, 103.53)
# img_std = (58.395, 57.12, 57.375)
img_std = (1., 1., 1.)
train_dataset = coco.CocoDataSet('dataset',
'train',
flip_ratio=0,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(800, 1024))
img, img_meta, bboxes, labels = train_dataset[0]
rgb_img = np.round(img + img_mean)
print('origin shape:', img_meta[:3])
print('scale shape:', img_meta[3:6])
print('pad shape:', img_meta[6:9])
print('scale factor:', img_meta[9:10])
print('flip :', img_meta[10:11])
print('newimg shape:', img.shape)
print('bboxes:', bboxes.shape)
print('labels:', labels)
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
print(tf.__version__)
assert tf.__version__.startswith('2.')
tf.random.set_seed(22)
np.random.seed(22)
img_mean = (123.675, 116.28, 103.53)
# img_std = (58.395, 57.12, 57.375)
img_std = (1., 1., 1.)
batch_size = 8
train_dataset = coco.CocoDataSet('/scratch/llong/datasets/coco2017/',
'train',
flip_ratio=0.5,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(800, 1216))
num_classes = len(train_dataset.get_categories())
train_generator = data_generator.DataGenerator(train_dataset)
train_tf_dataset = tf.data.Dataset.from_generator(
train_generator, (tf.float32, tf.float32, tf.float32, tf.int32))
train_tf_dataset = train_tf_dataset.batch(batch_size).prefetch(100).shuffle(
100)
# train_tf_dataset = train_tf_dataset.padded_batch(
# batch_size, padded_shapes=([None, None, None], [None], [None, None], [None]))
model = faster_rcnn.FasterRCNN(num_classes=num_classes)
optimizer = keras.optimizers.SGD(1e-3, momentum=0.9, nesterov=True)
# When you are finetuning with some pretrained model,
# you need to follow the mean and std values used for pretraining.
img_mean = (123.675, 116.28, 103.53)
img_std = (58.395, 57.12, 57.375)
epochs = 5
batch_size = 2
flip_ratio = 0
learning_rate = 1e-4
finetune = False
train_dataset = coco.CocoDataSet(
dataset_dir='./val2014',
subset='train',
annotation_dir='./annotations/sampled_ann_train.json',
flip_ratio=flip_ratio,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(800, 1216))
test_dataset = coco.CocoDataSet(
dataset_dir='./val2014',
subset='val',
annotation_dir='./annotations/sampled_ann_test.json',
flip_ratio=flip_ratio,
pad_mode='non-fixed',
mean=img_mean,
std=img_std,
scale=(800, 1216))
train_generator = data_generator.DataGenerator(train_dataset)
epochs = int(arg)
elif opt == '-c':
checkpoint = int(arg)
elif opt == '-n':
if int(arg) == 0:
img_mean = (0., 0., 0.)
img_std = (1., 1., 1.)
elif int(arg) == 1:
# Company Articles Dataset
img_mean = (0.9684, 0.9683, 0.9683)
img_std = (0.1502, 0.1505, 0.1505)
train_dataset = coco.CocoDataSet(dataset_dir='dataset',
subset='train',
flip_ratio=flip_ratio,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(800, 1216))
test_dataset = coco.CocoDataSet(dataset_dir='dataset',
subset='val',
flip_ratio=flip_ratio,
pad_mode='non-fixed',
mean=img_mean,
std=img_std,
scale=(800, 1216))
train_generator = data_generator.DataGenerator(train_dataset)
train_tf_dataset = tf.data.Dataset.from_generator(
train_generator, (tf.float32, tf.float32, tf.float32, tf.int32))
train_tf_dataset = train_tf_dataset.batch(batch_size).prefetch(100).shuffle(
# session = tf.Session(config=config)
print("Num GPUs Available: ",
len(tf.config.experimental.list_physical_devices('GPU')))
for gpu in tf.config.experimental.list_physical_devices('GPU'):
tf.config.experimental.set_memory_growth(gpu, True)
# load dataset
img_mean = (123.675, 116.28, 103.53)
# img_std = (58.395, 57.12, 57.375)
img_std = (1., 1., 1.)
train_dataset = coco.CocoDataSet('./COCO2017/',
'val',
flip_ratio=0.5,
pad_mode='fixed',
mean=img_mean,
std=img_std,
scale=(400, 512))
train_generator = data_generator.DataGenerator(train_dataset)
# display a sample
img, img_meta, bboxes, labels = train_dataset[0]
rgb_img = np.round(img + img_mean)
ori_img = get_original_image(img, img_meta, img_mean)
visualize.display_instances(rgb_img, bboxes, labels,
train_dataset.get_categories())
# load model
model = faster_rcnn.FasterRCNN(num_classes=len(train_dataset.get_categories()))