From 31fc66b795c0a57b8009d7b03f49f6cd099ceb29 Mon Sep 17 00:00:00 2001
From: natanielruiz <nataniel777@hotmail.com>
Date: 星期六, 23 九月 2017 12:07:48 +0800
Subject: [PATCH] Trying superres
---
code/train.py | 189 ++++++++++++++++++++++++++++++++++++-----------
1 files changed, 145 insertions(+), 44 deletions(-)
diff --git a/code/train.py b/code/train.py
index bef7aca..2f0cce3 100644
--- a/code/train.py
+++ b/code/train.py
@@ -33,6 +33,8 @@
default=0, type=int)
parser.add_argument('--num_epochs', dest='num_epochs', help='Maximum number of training epochs.',
default=5, type=int)
+ parser.add_argument('--num_epochs_ft', dest='num_epochs_ft', help='Maximum number of finetuning epochs.',
+ default=5, type=int)
parser.add_argument('--batch_size', dest='batch_size', help='Batch size.',
default=16, type=int)
parser.add_argument('--lr', dest='lr', help='Base learning rate.',
@@ -41,9 +43,13 @@
default='', type=str)
parser.add_argument('--filename_list', dest='filename_list', help='Path to text file containing relative paths for every example.',
default='', type=str)
-
+ parser.add_argument('--output_string', dest='output_string', help='String appended to output snapshots.', default = '', type=str)
+ parser.add_argument('--alpha', dest='alpha', help='Regression loss coefficient.',
+ default=0.001, type=float)
+ parser.add_argument('--iter_ref', dest='iter_ref', help='Number of iterative refinement passes.',
+ default=1, type=int)
+ parser.add_argument('--dataset', dest='dataset', help='Dataset type.', default='Pose_300W_LP', type=str)
args = parser.parse_args()
-
return args
def get_ignored_params(model):
@@ -51,25 +57,37 @@
b = []
b.append(model.conv1)
b.append(model.bn1)
+ for i in range(len(b)):
+ for module_name, module in b[i].named_modules():
+ if 'bn' in module_name:
+ module.eval()
+ for name, param in module.named_parameters():
+ yield param
+
+def get_non_ignored_params(model):
+ # Generator function that yields params that will be optimized.
+ b = []
b.append(model.layer1)
b.append(model.layer2)
b.append(model.layer3)
b.append(model.layer4)
for i in range(len(b)):
- for j in b[i].modules():
- for k in j.parameters():
- yield k
+ for module_name, module in b[i].named_modules():
+ if 'bn' in module_name:
+ module.eval()
+ for name, param in module.named_parameters():
+ yield param
-def get_non_ignored_params(model):
- # Generator function that yields params that will be optimized.
+def get_fc_params(model):
b = []
b.append(model.fc_yaw)
b.append(model.fc_pitch)
b.append(model.fc_roll)
+ b.append(model.fc_finetune)
for i in range(len(b)):
- for j in b[i].modules():
- for k in j.parameters():
- yield k
+ for module_name, module in b[i].named_modules():
+ for name, param in module.named_parameters():
+ yield param
def load_filtered_state_dict(model, snapshot):
# By user apaszke from discuss.pytorch.org
@@ -86,6 +104,7 @@
cudnn.enabled = True
num_epochs = args.num_epochs
+ num_epochs_ft = args.num_epochs_ft
batch_size = args.batch_size
gpu = args.gpu_id
@@ -95,67 +114,89 @@
# ResNet101 with 3 outputs
# model = hopenet.Hopenet(torchvision.models.resnet.Bottleneck, [3, 4, 23, 3], 66)
# ResNet50
- model = hopenet.Hopenet(torchvision.models.resnet.Bottleneck, [3, 4, 6, 3], 66)
+ model = hopenet.Hopenet(torchvision.models.resnet.Bottleneck, [3, 4, 6, 3], 66, args.iter_ref)
# ResNet18
# model = hopenet.Hopenet(torchvision.models.resnet.BasicBlock, [2, 2, 2, 2], 66)
load_filtered_state_dict(model, model_zoo.load_url(model_urls['resnet50']))
print 'Loading data.'
- transformations = transforms.Compose([transforms.Scale(224),transforms.RandomCrop(224),
- transforms.ToTensor()])
+ transformations = transforms.Compose([transforms.Scale(240),
+ transforms.RandomCrop(224), transforms.ToTensor(),
+ transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
- pose_dataset = datasets.300W_LP(args.data_dir, args.filename_list,
- transformations)
+ if args.dataset == 'Pose_300W_LP':
+ pose_dataset = datasets.Pose_300W_LP(args.data_dir, args.filename_list, transformations)
+ elif args.dataset == 'AFLW2000':
+ pose_dataset = datasets.AFLW2000(args.data_dir, args.filename_list, transformations)
+ elif args.dataset == 'BIWI':
+ pose_dataset = datasets.BIWI(args.data_dir, args.filename_list, transformations)
+ elif args.dataset == 'AFLW':
+ pose_dataset = datasets.AFLW(args.data_dir, args.filename_list, transformations)
+ elif args.dataset == 'AFLW_aug':
+ pose_dataset = datasets.AFLW_aug(args.data_dir, args.filename_list, transformations)
+ elif args.dataset == 'AFW':
+ pose_dataset = datasets.AFW(args.data_dir, args.filename_list, transformations)
+ else:
+ print 'Error: not a valid dataset name'
+ sys.exit()
train_loader = torch.utils.data.DataLoader(dataset=pose_dataset,
batch_size=batch_size,
shuffle=True,
num_workers=2)
model.cuda(gpu)
+ softmax = nn.Softmax()
criterion = nn.CrossEntropyLoss().cuda()
reg_criterion = nn.MSELoss().cuda()
# Regression loss coefficient
- alpha = 0.01
+ alpha = args.alpha
idx_tensor = [idx for idx in xrange(66)]
- idx_tensor = torch.FloatTensor(idx_tensor).cuda(gpu)
+ idx_tensor = Variable(torch.FloatTensor(idx_tensor)).cuda(gpu)
- optimizer = torch.optim.Adam([{'params': get_ignored_params(model), 'lr': args.lr},
- {'params': get_non_ignored_params(model), 'lr': args.lr * 10}],
- lr = args.lr)
+ optimizer = torch.optim.Adam([{'params': get_ignored_params(model), 'lr': 0},
+ {'params': get_non_ignored_params(model), 'lr': args.lr},
+ {'params': get_fc_params(model), 'lr': args.lr * 2}],
+ lr = args.lr)
print 'Ready to train network.'
+ print 'First phase of training.'
for epoch in range(num_epochs):
- for i, (images, labels, name) in enumerate(train_loader):
+ for i, (images, labels, cont_labels, name) in enumerate(train_loader):
images = Variable(images.cuda(gpu))
label_yaw = Variable(labels[:,0].cuda(gpu))
label_pitch = Variable(labels[:,1].cuda(gpu))
label_roll = Variable(labels[:,2].cuda(gpu))
+ label_angles = Variable(cont_labels[:,:3].cuda(gpu))
+ label_yaw_cont = Variable(cont_labels[:,0].cuda(gpu))
+ label_pitch_cont = Variable(cont_labels[:,1].cuda(gpu))
+ label_roll_cont = Variable(cont_labels[:,2].cuda(gpu))
+
optimizer.zero_grad()
model.zero_grad()
- yaw, pitch, roll = model(images)
+ pre_yaw, pre_pitch, pre_roll, angles = model(images)
# Cross entropy loss
- loss_yaw = criterion(yaw, label_yaw)
- loss_pitch = criterion(pitch, label_pitch)
- loss_roll = criterion(roll, label_roll)
+ loss_yaw = criterion(pre_yaw, label_yaw)
+ loss_pitch = criterion(pre_pitch, label_pitch)
+ loss_roll = criterion(pre_roll, label_roll)
# MSE loss
- yaw_predicted = F.softmax(yaw)
- pitch_predicted = F.softmax(pitch)
- roll_predicted = F.softmax(roll)
+ yaw_predicted = softmax(pre_yaw)
+ pitch_predicted = softmax(pre_pitch)
+ roll_predicted = softmax(pre_roll)
- yaw_predicted = torch.sum(yaw_predicted.data * idx_tensor, 1)
- pitch_predicted = torch.sum(pitch_predicted.data * idx_tensor, 1)
- roll_predicted = torch.sum(roll_predicted.data * idx_tensor, 1)
+ yaw_predicted = torch.sum(yaw_predicted * idx_tensor, 1) * 3 - 99
+ pitch_predicted = torch.sum(pitch_predicted * idx_tensor, 1) * 3 - 99
+ roll_predicted = torch.sum(roll_predicted * idx_tensor, 1) * 3 - 99
- loss_reg_yaw = reg_criterion(yaw_predicted, label_yaw.float())
- loss_reg_pitch = reg_criterion(pitch_predicted, label_pitch.float())
- loss_reg_roll = reg_criterion(roll_predicted, label_roll.float())
+ loss_reg_yaw = reg_criterion(yaw_predicted, label_yaw_cont)
+ loss_reg_pitch = reg_criterion(pitch_predicted, label_pitch_cont)
+ loss_reg_roll = reg_criterion(roll_predicted, label_roll_cont)
# Total loss
loss_yaw += alpha * loss_reg_yaw
@@ -167,21 +208,81 @@
torch.autograd.backward(loss_seq, grad_seq)
optimizer.step()
- # print ('Epoch [%d/%d], Iter [%d/%d] Losses: Yaw %.4f, Pitch %.4f, Roll %.4f'
- # %(epoch+1, num_epochs, i+1, len(pose_dataset)//batch_size, loss_yaw.data[0], loss_pitch.data[0], loss_roll.data[0]))
-
if (i+1) % 100 == 0:
print ('Epoch [%d/%d], Iter [%d/%d] Losses: Yaw %.4f, Pitch %.4f, Roll %.4f'
%(epoch+1, num_epochs, i+1, len(pose_dataset)//batch_size, loss_yaw.data[0], loss_pitch.data[0], loss_roll.data[0]))
- if epoch == 0:
- torch.save(model.state_dict(),
- 'output/snapshots/resnet50_iter_'+ str(i+1) + '.pkl')
+ # if epoch == 0:
+ # torch.save(model.state_dict(),
+ # 'output/snapshots/' + args.output_string + '_iter_'+ str(i+1) + '.pkl')
# Save models at numbered epochs.
- if epoch % 1 == 0 and epoch < num_epochs - 1:
+ if epoch % 1 == 0 and epoch < num_epochs:
print 'Taking snapshot...'
torch.save(model.state_dict(),
- 'output/snapshots/resnet50_epoch_'+ str(epoch+1) + '.pkl')
+ 'output/snapshots/' + args.output_string + '_epoch_'+ str(epoch+1) + '.pkl')
- # Save the final Trained Model
- torch.save(model.state_dict(), 'output/snapshots/resnet50_epoch' + str(epoch+1) + '.pkl')
+ print 'Second phase of training (finetuning layer).'
+ for epoch in range(num_epochs_ft):
+ for i, (images, labels, cont_labels, name) in enumerate(train_loader):
+ images = Variable(images.cuda(gpu))
+ label_yaw = Variable(labels[:,0].cuda(gpu))
+ label_pitch = Variable(labels[:,1].cuda(gpu))
+ label_roll = Variable(labels[:,2].cuda(gpu))
+
+ label_angles = Variable(cont_labels[:,:3].cuda(gpu))
+ label_yaw_cont = Variable(cont_labels[:,0].cuda(gpu))
+ label_pitch_cont = Variable(cont_labels[:,1].cuda(gpu))
+ label_roll_cont = Variable(cont_labels[:,2].cuda(gpu))
+
+ optimizer.zero_grad()
+ model.zero_grad()
+
+ pre_yaw, pre_pitch, pre_roll, angles = model(images)
+
+ # Cross entropy loss
+ loss_yaw = criterion(pre_yaw, label_yaw)
+ loss_pitch = criterion(pre_pitch, label_pitch)
+ loss_roll = criterion(pre_roll, label_roll)
+
+ # MSE loss
+ yaw_predicted = softmax(pre_yaw)
+ pitch_predicted = softmax(pre_pitch)
+ roll_predicted = softmax(pre_roll)
+
+ yaw_predicted = torch.sum(yaw_predicted * idx_tensor, 1) * 3 - 99
+ pitch_predicted = torch.sum(pitch_predicted * idx_tensor, 1) * 3 - 99
+ roll_predicted = torch.sum(roll_predicted * idx_tensor, 1) * 3 - 99
+
+ loss_reg_yaw = reg_criterion(yaw_predicted, label_yaw_cont)
+ loss_reg_pitch = reg_criterion(pitch_predicted, label_pitch_cont)
+ loss_reg_roll = reg_criterion(roll_predicted, label_roll_cont)
+
+ # Total loss
+ loss_yaw += alpha * loss_reg_yaw
+ loss_pitch += alpha * loss_reg_pitch
+ loss_roll += alpha * loss_reg_roll
+
+ # Finetuning loss
+ loss_seq = [loss_yaw, loss_pitch, loss_roll]
+ for idx in xrange(1,len(angles)):
+ label_angles_residuals = label_angles - angles[0] * 3 - 99
+ label_angles_residuals = label_angles_residuals.detach()
+ loss_angles = reg_criterion(angles[idx], label_angles_residuals)
+ loss_seq.append(loss_angles)
+
+ grad_seq = [torch.Tensor(1).cuda(gpu) for _ in range(len(loss_seq))]
+ torch.autograd.backward(loss_seq, grad_seq)
+ optimizer.step()
+
+ if (i+1) % 100 == 0:
+ print ('Epoch [%d/%d], Iter [%d/%d] Losses: pre-yaw %.4f, pre-pitch %.4f, pre-roll %.4f, finetuning %.4f'
+ %(epoch+1, num_epochs_ft, i+1, len(pose_dataset)//batch_size, loss_yaw.data[0], loss_pitch.data[0], loss_roll.data[0], loss_angles.data[0]))
+ # if epoch == 0:
+ # torch.save(model.state_dict(),
+ # 'output/snapshots/' + args.output_string + '_iter_'+ str(i+1) + '.pkl')
+
+ # Save models at numbered epochs.
+ if epoch % 1 == 0 and epoch < num_epochs_ft:
+ print 'Taking snapshot...'
+ torch.save(model.state_dict(),
+ 'output/snapshots/' + args.output_string + '_epoch_'+ str(num_epochs+epoch+1) + '.pkl')
--
Gitblit v1.8.0