def get_data_loader(opt, batch_size, num_workers, logger):
data_dir = opt.data_dir
normalize = video.VideoNormalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
scale_ratios = [1.0, 0.875, 0.75, 0.66]
input_size = opt.input_size
def batch_fn(batch, ctx):
if opt.num_segments > 1:
data = split_and_load(batch[0], ctx_list=ctx, batch_axis=0, even_split=False, multiplier=opt.num_segments)
else:
data = split_and_load(batch[0], ctx_list=ctx, batch_axis=0, even_split=False)
label = split_and_load(batch[1], ctx_list=ctx, batch_axis=0, even_split=False)
return data, label
transform_train = transforms.Compose([
video.VideoMultiScaleCrop(size=(input_size, input_size), scale_ratios=scale_ratios),
video.VideoRandomHorizontalFlip(),
video.VideoToTensor(),
normalize
])
transform_test = transforms.Compose([
video.VideoCenterCrop(size=input_size),
video.VideoToTensor(),
normalize
])
train_dataset = ucf101.classification.UCF101(setting=opt.train_list, root=data_dir, train=True,
new_width=opt.new_width, new_height=opt.new_height,
target_width=input_size, target_height=input_size,
num_segments=opt.num_segments, transform=transform_train)
val_dataset = ucf101.classification.UCF101(setting=opt.val_list, root=data_dir, train=False,
new_width=opt.new_width, new_height=opt.new_height,
target_width=input_size, target_height=input_size,
num_segments=opt.num_segments, transform=transform_test)
logger.info('Load %d training samples and %d validation samples.' % (len(train_dataset), len(val_dataset)))
if opt.num_segments > 1:
train_data = gluon.data.DataLoader(train_dataset, batch_size=batch_size, shuffle=True, num_workers=num_workers, batchify_fn=tsn_mp_batchify_fn)
val_data = gluon.data.DataLoader(val_dataset, batch_size=batch_size, shuffle=False, num_workers=num_workers, batchify_fn=tsn_mp_batchify_fn)
else:
train_data = gluon.data.DataLoader(train_dataset, batch_size=batch_size, shuffle=True, num_workers=num_workers)
val_data = gluon.data.DataLoader(val_dataset, batch_size=batch_size, shuffle=False, num_workers=num_workers)
return train_data, val_data, batch_fn
################################################################
# Data Augmentation and Data Loader
# ---------------------------------
#
# Data augmentation for video is different from image. For example, if you
# want to randomly crop a video sequence, you need to make sure all the video
# frames in this sequence undergo the same cropping process. We provide a
# new set of transformation functions, working with multiple images.
# Please checkout the `video.py <../../../gluoncv/data/transforms/video.py>`_ for more details.
# Most video data augmentation strategies used here are introduced in [Wang15]_.
transform_train = transforms.Compose([
# Fix the input video frames size as 256×340 and randomly sample the cropping width and height from
# {256,224,192,168}. After that, resize the cropped regions to 224 × 224.
video.VideoMultiScaleCrop(size=(224, 224),
scale_ratios=[1.0, 0.875, 0.75, 0.66]),
# Randomly flip the video frames horizontally
video.VideoRandomHorizontalFlip(),
# Transpose the video frames from height*width*num_channels to num_channels*height*width
# and map values from [0, 255] to [0,1]
video.VideoToTensor(),
# Normalize the video frames with mean and standard deviation calculated across all images
video.VideoNormalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
])
##################################################################
# With the transform functions, we can define data loaders for our
# training datasets.
# Batch Size for Each GPU
per_device_batch_size = 25
