import matplotlib.pyplot as plt
plt.figure(figsize=figsize)
plt.imshow(x, interpolation='nearest', cmap='jet')
if title:
plt.title(title)
if cbar:
plt.colorbar()
plt.show()
if __name__ == '__main__':
args = parse_args()
model1 = DnCNN()
model2 = DecomposeNet(image_channels=5, n_block=32)
model3
# model = torch.load(os.path.join(args.model_dir, args.model_name))
model.load_state_dict(
torch.load(os.path.join(args.model_dir, args.model_name)))
decompose_model.load_state_dict(
torch.load(os.path.join(args.decom_model_dir, args.decom_model_name)))
# model = torch.load(os.path.join(args.model_dir, args.model_name))
log('load trained model')
# params = model.state_dict()
# print(params.values())
# print(params.keys())
#
# for key, value in params.items():
import matplotlib.pyplot as plt
plt.figure(figsize=figsize)
plt.imshow(x, interpolation='nearest', cmap='jet')
if title:
plt.title(title)
if cbar:
plt.colorbar()
plt.show()
if __name__ == '__main__':
args = parse_args()
model = DnCNN()
decompose_model = DecomposeNet(image_channels=5, n_block=32)
if not os.path.exists(
os.path.join(args.decom_model_dir, args.decom_model_name)):
decompose_model = torch.load(os.path.join(args.model_dir, 'model.pth'))
# load weights into new model
log('load trained model on Train400 dataset by kai')
else:
model = torch.load(os.path.join(args.model_dir, args.model_name))
# model.load_state_dict(torch.load(os.path.join(args.model_dir, args.model_name)))
decompose_model.load_state_dict(
torch.load(
os.path.join(args.decom_model_dir, args.decom_model_name)))
# model = torch.load(os.path.join(args.model_dir, args.model_name))
log('load trained model')
dim=1)
Low_output = torch.cat(
(torch.tensor(np.real(np.fft.ifft2(Low_freq))).unsqueeze(1).float(),
torch.tensor(np.imag(np.fft.ifft2(Low_freq))).unsqueeze(1).float()),
dim=1)
return High_output, Low_output
if not os.path.exists(save_dir):
os.mkdir(save_dir)
if __name__ == '__main__':
# model selection
print('===> Building model')
model_hf = DecomposeNet(image_channels=2, n_block=32)
model_lf = DecomposeNet(image_channels=2, n_block=32)
initial_epoch = findLastCheckpoint(
save_dir=save_dir) # load the last model in matconvnet style
if initial_epoch > 0:
print('resuming by loading epoch %03d' % initial_epoch)
model_hf.load_state_dict(
torch.load(
os.path.join(save_dir, 'high_model_%03d.pth' % initial_epoch)))
model_lf.load_state_dict(
torch.load(
os.path.join(save_dir, 'low_model_%03d.pth' % initial_epoch)))
# model_hf = torch.load(os.path.join(save_dir, 'high_model_%03d.pth' % initial_epoch))
# model_lf = torch.load(os.path.join(save_dir, 'low_model_%03d.pth' % initial_epoch))
model_hf.train()
batch_size = args.batch_size
cuda = torch.cuda.is_available()
n_epoch = args.epoch
sigma = args.sigma
save_dir = args.save_dir
# save_dir = os.path.join('models', args.model+'_' + 'sigma' + str(sigma))
if not os.path.exists(save_dir):
os.mkdir(save_dir)
if __name__ == '__main__':
# model selection
print('===> Building model')
model = DecomposeNet()
initial_epoch = findLastCheckpoint(save_dir=save_dir) # load the last model in matconvnet style
if initial_epoch > 0:
print('resuming by loading epoch %03d' % initial_epoch)
# model.load_state_dict(torch.load(os.path.join(save_dir, 'model_%03d.pth' % initial_epoch)))
# model = torch.load(os.path.join(save_dir, 'model_%03d.pth' % initial_epoch))
model.train()
# criterion = nn.MSELoss(reduction = 'sum') # PyTorch 0.4.1
# criterion = sum_squared_error()
criterion = nn.MSELoss()
if cuda:
model = model.cuda()
# device_ids = [0]
# model = nn.DataParallel(model, device_ids=device_ids).cuda()
dim=1)
Low_output = torch.cat(
(torch.tensor(np.real(np.fft.ifft2(Low_freq))).unsqueeze(1).float(),
torch.tensor(np.imag(np.fft.ifft2(Low_freq))).unsqueeze(1).float()),
dim=1)
return High_output, Low_output
if not os.path.exists(save_dir):
os.mkdir(save_dir)
if __name__ == '__main__':
# model selection
print('===> Building model')
model = DecomposeNet(image_channels=4, n_block=32)
# model_lf = DecomposeNet(image_channels=2, n_block=32)
initial_epoch = findLastCheckpoint(
save_dir=save_dir) # load the last model in matconvnet style
if initial_epoch > 0:
print('resuming by loading epoch %03d' % initial_epoch)
model.load_state_dict(
torch.load(os.path.join(save_dir,
'model_%03d.pth' % initial_epoch)))
# model_hf = torch.load(os.path.join(save_dir, 'high_model_%03d.pth' % initial_epoch))
# model_lf = torch.load(os.path.join(save_dir, 'low_model_%03d.pth' % initial_epoch))
model.train()
# criterion = nn.MSELoss(reduction = 'sum') # PyTorch 0.4.1
dim=1)
Low_output = torch.cat(
(torch.tensor(np.real(np.fft.ifft2(Low_freq))).unsqueeze(1).float(),
torch.tensor(np.imag(np.fft.ifft2(Low_freq))).unsqueeze(1).float()),
dim=1)
return High_output, Low_output
if not os.path.exists(save_dir):
os.mkdir(save_dir)
if __name__ == '__main__':
# model selection
print('===> Building model')
decompose_model = DecomposeNet(image_channels=5, n_block=32)
compose_model = ComposeNet(n_block=32)
initial_epoch = findLastCheckpoint(
save_dir=save_dir) # load the last model in matconvnet style
# initial_epoch = 1
if initial_epoch > 0:
print('resuming by loading epoch %03d' % initial_epoch)
decompose_model.load_state_dict(
torch.load(
os.path.join(save_dir,
'decom_model_%03d.pth' % initial_epoch)))
# model = torch.load(os.path.join(save_dir, 'model_%03d.pth' % initial_epoch))
# criterion = nn.MSELoss(reduction = 'sum') # PyTorch 0.4.1
# criterion = sum_squared_error()
import matplotlib.pyplot as plt
plt.figure(figsize=figsize)
plt.imshow(x, interpolation='nearest', cmap='jet')
if title:
plt.title(title)
if cbar:
plt.colorbar()
plt.show()
if __name__ == '__main__':
args = parse_args()
model = DecomposeNet(image_channels=4, n_block=32)
if not os.path.exists(os.path.join(args.model_dir, args.model_name)):
model = torch.load(os.path.join(args.model_dir, 'model.pth'))
# load weights into new model
log('load trained model on Train400 dataset by kai')
else:
model.load_state_dict(torch.load(os.path.join(args.model_dir, args.model_name)))
# model = torch.load(os.path.join(args.model_dir, args.model_name))
log('load trained model')
# params = model.state_dict()
# print(params.values())
# print(params.keys())
#
# for key, value in params.items():
import matplotlib.pyplot as plt
plt.figure(figsize=figsize)
plt.imshow(x, interpolation='nearest', cmap='gray')
if title:
plt.title(title)
if cbar:
plt.colorbar()
plt.show()
if __name__ == '__main__':
args = parse_args()
model = DecomposeNet()
if not os.path.exists(os.path.join(args.model_dir, args.model_name)):
model = torch.load(os.path.join(args.model_dir, 'model.pth'))
# load weights into new model
log('load trained model on Train400 dataset by kai')
else:
model.load_state_dict(torch.load(os.path.join(args.model_dir, args.model_name)))
# model = torch.load(os.path.join(args.model_dir, args.model_name))
log('load trained model')
# params = model.state_dict()
# print(params.values())
# print(params.keys())
#
# for key, value in params.items():