Main function to carry out the training loop.
This function creates the generator and data loaders. Then, it trains the generator.
"""
device = 'cuda' if torch.cuda.is_available() else 'cpu'
RANDOM_ALL = True
PRECROP = True if DATASET.lower() == 'ntu' else False
VP_VALUE_COUNT = 1 if DATASET.lower() == 'ntu' else 3
CLOSE_VIEWS = True if DATASET.lower() == 'panoptic' else False
if DATASET.lower() == 'ntu':
data_root_dir, test_split, param_file, weights_path, output_video_dir = ntu_config(
)
# generator
model = FullNetwork(vp_value_count=VP_VALUE_COUNT,
output_shape=(BATCH_SIZE, CHANNELS, FRAMES, HEIGHT,
WIDTH))
model.load_state_dict(torch.load(weights_path))
model = model.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(model)
cudnn.benchmark = True
criterion = nn.MSELoss()
# data
testset = NTUDataset(root_dir=data_root_dir,
data_file=test_split,
param_file=param_file,
resize_height=HEIGHT,
"""
Main function to carry out the training loop.
This function creates the generator and data loaders. Then, it trains the generator.
"""
device = 'cuda' if torch.cuda.is_available() else 'cpu'
RANDOM_ALL = True
PRECROP = True if DATASET.lower() == 'ntu' else False
VP_VALUE_COUNT = 1 if DATASET.lower() == 'ntu' else 3
CLOSE_VIEWS = True if DATASET.lower() == 'panoptic' else False
if DATASET.lower() == 'ntu':
data_root_dir, train_split, test_split, param_file, weight_file = ntu_config()
# generator
model = FullNetwork(vp_value_count=VP_VALUE_COUNT,
output_shape=(BATCH_SIZE, CHANNELS, FRAMES, HEIGHT, WIDTH),
stdev=STDEV)
model = model.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(model)
cudnn.benchmark = True
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=LR)
# data
trainset = NTUDataset(root_dir=data_root_dir, data_file=train_split, param_file=param_file,
resize_height=HEIGHT, resize_width=WIDTH,
clip_len=FRAMES, skip_len=SKIP_LEN,
random_all=RANDOM_ALL, precrop=PRECROP)
Main function to carry out the training loop.
This function creates the generator and data loaders. Then, it trains the generator.
"""
device = 'cuda' if torch.cuda.is_available() else 'cpu'
RANDOM_ALL = True
PRECROP = True if DATASET.lower() == 'ntu' else False
VP_VALUE_COUNT = 1 if DATASET.lower() == 'ntu' else 3
CLOSE_VIEWS = True if DATASET.lower() == 'panoptic' else False
if DATASET.lower() == 'ntu':
data_root_dir, train_split, test_split, param_file, weight_file = ntu_config(
)
# generator
model = FullNetwork(vp_value_count=VP_VALUE_COUNT,
stdev=STDEV,
output_shape=(BATCH_SIZE, CHANNELS, FRAMES, HEIGHT,
WIDTH))
if pretrained:
model.load_state_dict(torch.load(pretrained_weights))
model = model.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(model)
cudnn.benchmark = True
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=LR)
# data
trainset = NTUDataset(root_dir=data_root_dir,
data_file=train_split,
if __name__ == '__main__':
"""
Main function to carry out the training loop.
This function creates the generator and data loaders. Then, it trains the generator.
"""
device = 'cuda' if torch.cuda.is_available() else 'cpu'
RANDOM_ALL = True
PRECROP = True if DATASET.lower() == 'ntu' else False
VP_VALUE_COUNT = 1 if DATASET.lower() == 'ntu' else 3
CLOSE_VIEWS = True if DATASET.lower() == 'panoptic' else False
# vgg_weights_path, i3d_weights_path = pretrained_weights_config()
# generator
generator = FullNetwork(vp_value_count=VP_VALUE_COUNT, stdev=STDEV,
output_shape=(BATCH_SIZE, CHANNELS, FRAMES, HEIGHT, WIDTH))
# pretrained=True, vgg_weights_path=vgg_weights_path, i3d_weights_path=i3d_weights_path)
generator = generator.to(device)
# discriminator
discriminator = Discriminator(in_channels=3)
discriminator = discriminator.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(generator)
cudnn.benchmark = True
# Loss functions
criterion = nn.MSELoss()
adversarial_loss = nn.BCELoss()
# categorical_loss = torch.nn.CrossEntropyLoss()
# continuous_loss = torch.nn.MSELoss()
if __name__ == '__main__':
"""
Main function to carry out the training loop.
This function creates the model and data loaders. Then, it trains the model.
"""
device = 'cuda' if torch.cuda.is_available() else 'cpu'
RANDOM_ALL = True
PRECROP = True if DATASET.lower() == 'ntu' else False
if DATASET.lower() == 'ntu':
data_root_dir, train_split, test_split, param_file, weight_file = ntu_config()
# model
model = FullNetwork(vp_value_count=1, output_shape=(BATCH_SIZE, CHANNELS, FRAMES, HEIGHT, WIDTH))
model = model.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(model)
cudnn.benchmark = True
criterion = nn.MSELoss()
optimizer = optim.Adam(model.parameters(), lr=LR)
# data
trainset = NTUDataset(root_dir=data_root_dir, data_file=train_split, param_file=param_file,
resize_height=HEIGHT, resize_width=WIDTH,
clip_len=FRAMES, skip_len=SKIP_LEN,
random_all=RANDOM_ALL, precrop=PRECROP)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=BATCH_SIZE, shuffle=True, num_workers=2)
This function creates the generator and data loaders. Then, it trains the generator.
"""
device = 'cuda' if torch.cuda.is_available() else 'cpu'
RANDOM_ALL = True
PRECROP = True if DATASET.lower() == 'ntu' else False
VP_VALUE_COUNT = 1 if DATASET.lower() == 'ntu' else 3
CLOSE_VIEWS = True if DATASET.lower() == 'panoptic' else False
vgg_weights_path, i3d_weights_path, gen_weights_path, disc_weights_path = pretrained_weights_config(
)
# generator
generator = FullNetwork(vp_value_count=VP_VALUE_COUNT,
stdev=STDEV,
output_shape=(BATCH_SIZE, CHANNELS, FRAMES, HEIGHT,
WIDTH),
pretrained=True,
vgg_weights_path=vgg_weights_path,
i3d_weights_path=i3d_weights_path)
if GEN_PRETRAINED:
generator.load_state_dict(torch.load(gen_weights_path))
generator = generator.to(device)
# discriminator
discriminator = Discriminator(in_channels=3,
pretrained=GEN_PRETRAINED,
weights_path=disc_weights_path)
discriminator = discriminator.to(device)
if device == 'cuda':
net = torch.nn.DataParallel(generator)
cudnn.benchmark = True
