From ec99c6649af6bdbd3c836f20cdc81170e7045cc8 Mon Sep 17 00:00:00 2001
From: natanielruiz <nataniel777@hotmail.com>
Date: 星期四, 14 九月 2017 10:06:48 +0800
Subject: [PATCH] Training hopenet and normal for different alpha values on AFLW
---
code/train.py | 2
/dev/null | 205 ---------------------------------------------------
code/old/test_shape.py | 0
code/old/train_shape.py | 0
code/train_preangles.py | 2
code/old/test_old.py | 0
code/test_preangles.py | 22 -----
7 files changed, 2 insertions(+), 229 deletions(-)
diff --git a/code/batch_testing/batch_testing_AFLW_preangles.py b/code/batch_testing/batch_testing_AFLW_preangles.py
deleted file mode 100644
index e0c170e..0000000
--- a/code/batch_testing/batch_testing_AFLW_preangles.py
+++ /dev/null
@@ -1,147 +0,0 @@
-import numpy as np
-import torch
-import torch.nn as nn
-from torch.autograd import Variable
-from torch.utils.data import DataLoader
-from torchvision import transforms
-import torch.backends.cudnn as cudnn
-import torchvision
-import torch.nn.functional as F
-
-import cv2
-import matplotlib.pyplot as plt
-import sys
-import os
-import argparse
-
-import datasets
-import hopenet
-import utils
-
-import glob
-
-def parse_args():
- """Parse input arguments."""
- parser = argparse.ArgumentParser(description='Head pose estimation using the Hopenet network.')
- parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]',
- default=0, type=int)
- parser.add_argument('--data_dir', dest='data_dir', help='Directory path for data.',
- 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('--snapshot_folder', dest='snapshot_folder', help='Name of model snapshot folder.',
- default='', type=str)
- parser.add_argument('--batch_size', dest='batch_size', help='Batch size.',
- default=1, type=int)
- parser.add_argument('--save_viz', dest='save_viz', help='Save images with pose cube.',
- default=False, type=bool)
-
- args = parser.parse_args()
-
- return args
-
-if __name__ == '__main__':
- args = parse_args()
-
- cudnn.enabled = True
- gpu = args.gpu_id
-
- # 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)
- # ResNet18
- # model = hopenet.Hopenet(torchvision.models.resnet.BasicBlock, [2, 2, 2, 2], 66)
-
- print 'Loading snapshot list.'
- # Load snapshot
- snapshot_list = sorted(glob.glob(os.path.join(args.snapshot_folder, '*.pkl')))
-
- print 'Loading data.'
-
- # transformations = transforms.Compose([transforms.Scale(224),
- # transforms.RandomCrop(224), transforms.ToTensor()])
-
- transformations = transforms.Compose([transforms.Scale(224),
- transforms.RandomCrop(224), transforms.ToTensor(),
- transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
-
- pose_dataset = datasets.AFLW(args.data_dir, args.filename_list,
- transformations)
- test_loader = torch.utils.data.DataLoader(dataset=pose_dataset,
- batch_size=args.batch_size,
- num_workers=2)
-
- model.cuda(gpu)
-
- print 'Ready to test network.'
-
- output_file_name = args.snapshot_folder.split('/')[-1] + '_AFLW_preangles.txt'
- txt_output = open(os.join('output/batch_snapshots', output_file_name), 'w')
-
- for snapshot_path in snapshot_list:
- snapshot_name = snapshot_path.split('/')[-1].split('.')[0]
- print 'Loading snapshot ' + snapshot_name
-
- saved_state_dict = torch.load(snapshot_path)
- model.load_state_dict(saved_state_dict)
-
- # Test the Model
- model.eval() # Change model to 'eval' mode (BN uses moving mean/var).
- total = 0
- n_margins = 20
- yaw_correct = np.zeros(n_margins)
- pitch_correct = np.zeros(n_margins)
- roll_correct = np.zeros(n_margins)
-
- idx_tensor = [idx for idx in xrange(66)]
- idx_tensor = torch.FloatTensor(idx_tensor).cuda(gpu)
-
- yaw_error = .0
- pitch_error = .0
- roll_error = .0
-
- l1loss = torch.nn.L1Loss(size_average=False)
-
- for i, (images, labels, name) in enumerate(test_loader):
- images = Variable(images).cuda(gpu)
- total += labels.size(0)
- label_yaw = labels[:,0].float()
- label_pitch = labels[:,1].float()
- label_roll = labels[:,2].float()
-
- yaw, pitch, roll, angles = model(images)
-
- # Binned predictions
- _, yaw_bpred = torch.max(yaw.data, 1)
- _, pitch_bpred = torch.max(pitch.data, 1)
- _, roll_bpred = torch.max(roll.data, 1)
-
- # Continuous predictions
- yaw_predicted = utils.softmax_temperature(yaw.data, 1)
- pitch_predicted = utils.softmax_temperature(pitch.data, 1)
- roll_predicted = utils.softmax_temperature(roll.data, 1)
-
- yaw_predicted = torch.sum(yaw_predicted * idx_tensor, 1).cpu()
- pitch_predicted = torch.sum(pitch_predicted * idx_tensor, 1).cpu()
- roll_predicted = torch.sum(roll_predicted * idx_tensor, 1).cpu()
-
- # Mean absolute error
- yaw_error += torch.sum(torch.abs(yaw_predicted - label_yaw) * 3)
- pitch_error += torch.sum(torch.abs(pitch_predicted - label_pitch) * 3)
- roll_error += torch.sum(torch.abs(roll_predicted - label_roll) * 3)
-
- if args.save_viz:
- name = name[0]
- cv2_img = cv2.imread(os.path.join(args.data_dir, name + '.jpg'))
- utils.plot_pose_cube(cv2_img, yaw_predicted[0] * 3 - 99, pitch_predicted[0] * 3 - 99, roll_predicted[0] * 3 - 99)
- cv2.imwrite(os.path.join('output/images', name + '.jpg'), cv2_img)
-
- print('Test error in degrees of the model on the ' + str(total) +
- ' test images. Yaw: %.4f, Pitch: %.4f, Roll: %.4f' % (yaw_error / total,
- pitch_error / total, roll_error / total))
- txt_output.write('Test error in degrees of model ' + snapshot_name + ' on the ' + str(total) +
- ' test images. Yaw: %.4f, Pitch: %.4f, Roll: %.4f \n' % (yaw_error / total,
- pitch_error / total, roll_error / total))
-
- txt_output.close()
diff --git a/code/test_old.py b/code/old/test_old.py
similarity index 100%
rename from code/test_old.py
rename to code/old/test_old.py
diff --git a/code/test_shape.py b/code/old/test_shape.py
similarity index 100%
rename from code/test_shape.py
rename to code/old/test_shape.py
diff --git a/code/train_shape.py b/code/old/train_shape.py
similarity index 100%
rename from code/train_shape.py
rename to code/old/train_shape.py
diff --git a/code/test_preangles.py b/code/test_preangles.py
index 7cf8ebb..1203578 100644
--- a/code/test_preangles.py
+++ b/code/test_preangles.py
@@ -60,9 +60,6 @@
print 'Loading data.'
- # transformations = transforms.Compose([transforms.Scale(224),
- # transforms.RandomCrop(224), transforms.ToTensor()])
-
transformations = transforms.Compose([transforms.Scale(224),
transforms.CenterCrop(224), transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
@@ -90,10 +87,6 @@
# Test the Model
model.eval() # Change model to 'eval' mode (BN uses moving mean/var).
total = 0
- n_margins = 20
- yaw_correct = np.zeros(n_margins)
- pitch_correct = np.zeros(n_margins)
- roll_correct = np.zeros(n_margins)
idx_tensor = [idx for idx in xrange(66)]
idx_tensor = torch.FloatTensor(idx_tensor).cuda(gpu)
@@ -132,29 +125,14 @@
pitch_error += torch.sum(torch.abs(pitch_predicted - label_pitch) * 3)
roll_error += torch.sum(torch.abs(roll_predicted - label_roll) * 3)
- # Binned Accuracy
- # for er in xrange(n_margins):
- # yaw_bpred[er] += (label_yaw[0] in range(yaw_bpred[0,0] - er, yaw_bpred[0,0] + er + 1))
- # pitch_bpred[er] += (label_pitch[0] in range(pitch_bpred[0,0] - er, pitch_bpred[0,0] + er + 1))
- # roll_bpred[er] += (label_roll[0] in range(roll_bpred[0,0] - er, roll_bpred[0,0] + er + 1))
-
- # print label_yaw[0], yaw_bpred[0,0]
-
# Save images with pose cube.
# TODO: fix for larger batch size
if args.save_viz:
name = name[0]
cv2_img = cv2.imread(os.path.join(args.data_dir, name + '.jpg'))
- #print os.path.join('output/images', name + '.jpg')
- #print label_yaw[0] * 3 - 99, label_pitch[0] * 3 - 99, label_roll[0] * 3 - 99
- #print yaw_predicted * 3 - 99, pitch_predicted * 3 - 99, roll_predicted * 3 - 99
utils.plot_pose_cube(cv2_img, yaw_predicted[0] * 3 - 99, pitch_predicted[0] * 3 - 99, roll_predicted[0] * 3 - 99)
cv2.imwrite(os.path.join('output/images', name + '.jpg'), cv2_img)
print('Test error in degrees of the model on the ' + str(total) +
' test images. Yaw: %.4f, Pitch: %.4f, Roll: %.4f' % (yaw_error / total,
pitch_error / total, roll_error / total))
-
- # Binned accuracy
- # for idx in xrange(len(yaw_correct)):
- # print yaw_correct[idx] / total, pitch_correct[idx] / total, roll_correct[idx] / total
diff --git a/code/train.py b/code/train.py
index 03d5cf5..fd0735a 100644
--- a/code/train.py
+++ b/code/train.py
@@ -251,7 +251,7 @@
loss_pitch += alpha * loss_reg_pitch
loss_roll += alpha * loss_reg_roll
- loss_yaw *= 0.35
+ loss_yaw *= 1
# Finetuning loss
loss_seq = [loss_yaw, loss_pitch, loss_roll]
diff --git a/code/train_AFLW.py b/code/train_AFLW.py
deleted file mode 100644
index f355f63..0000000
--- a/code/train_AFLW.py
+++ /dev/null
@@ -1,197 +0,0 @@
-import numpy as np
-import torch
-import torch.nn as nn
-from torch.autograd import Variable
-from torch.utils.data import DataLoader
-from torchvision import transforms
-import torchvision
-import torch.backends.cudnn as cudnn
-import torch.nn.functional as F
-
-import cv2
-import matplotlib.pyplot as plt
-import sys
-import os
-import argparse
-
-import datasets
-import hopenet
-import torch.utils.model_zoo as model_zoo
-
-model_urls = {
- 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth',
- 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth',
- 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth',
- 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth',
- 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth',
-}
-
-def parse_args():
- """Parse input arguments."""
- parser = argparse.ArgumentParser(description='Head pose estimation using the Hopenet network.')
- parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]',
- 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('--batch_size', dest='batch_size', help='Batch size.',
- default=16, type=int)
- parser.add_argument('--lr', dest='lr', help='Base learning rate.',
- default=0.001, type=float)
- parser.add_argument('--data_dir', dest='data_dir', help='Directory path for data.',
- 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('--finetune', dest='finetune', help='Boolean: finetune or from Imagenet pretrain.',
- default=False, type=bool)
- parser.add_argument('--snapshot', dest='snapshot', help='Path to finetune snapshot.',
- default='', type=str)
-
- args = parser.parse_args()
-
- return args
-
-def get_ignored_params(model):
- # Generator function that yields ignored params.
- b = []
- b.append(model.conv1)
- b.append(model.bn1)
- 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
-
-def get_non_ignored_params(model):
- # Generator function that yields params that will be optimized.
- b = []
- b.append(model.fc_yaw)
- b.append(model.fc_pitch)
- b.append(model.fc_roll)
- for i in range(len(b)):
- for j in b[i].modules():
- for k in j.parameters():
- yield k
-
-def load_filtered_state_dict(model, snapshot):
- # By user apaszke from discuss.pytorch.org
- model_dict = model.state_dict()
- # 1. filter out unnecessary keys
- snapshot = {k: v for k, v in snapshot.items() if k in model_dict}
- # 2. overwrite entries in the existing state dict
- model_dict.update(snapshot)
- # 3. load the new state dict
- model.load_state_dict(model_dict)
-
-if __name__ == '__main__':
- args = parse_args()
-
- cudnn.enabled = True
- num_epochs = args.num_epochs
- batch_size = args.batch_size
- gpu = args.gpu_id
-
- if not os.path.exists('output/snapshots'):
- os.makedirs('output/snapshots')
-
- # 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)
- # ResNet18
- # model = hopenet.Hopenet(torchvision.models.resnet.BasicBlock, [2, 2, 2, 2], 66)
-
- if args.finetune:
- model.load_state_dict(torch.load(args.snapshot))
- else:
- 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()])
-
- pose_dataset = datasets.AFLW(args.data_dir, args.filename_list,
- transformations)
- train_loader = torch.utils.data.DataLoader(dataset=pose_dataset,
- batch_size=batch_size,
- shuffle=True,
- num_workers=2)
-
- model.cuda(gpu)
- criterion = nn.CrossEntropyLoss().cuda(gpu)
- reg_criterion = nn.MSELoss().cuda(gpu)
- # Regression loss coefficient
- alpha = 0.1
-
- idx_tensor = [idx for idx in xrange(66)]
- idx_tensor = 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.SGD([{'params': get_ignored_params(model), 'lr': args.lr},
- # {'params': get_non_ignored_params(model), 'lr': args.lr * 10}],
- # lr = args.lr, momentum = 0.9)
-
- print 'Ready to train network.'
-
- for epoch in range(num_epochs):
- for i, (images, 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))
-
- optimizer.zero_grad()
-
- yaw, pitch, roll = model(images)
-
- # Cross entropy loss
- loss_yaw = criterion(yaw, label_yaw)
- loss_pitch = criterion(pitch, label_pitch)
- loss_roll = criterion(roll, label_roll)
-
- # MSE loss
- yaw_predicted = F.softmax(yaw)
- pitch_predicted = F.softmax(pitch)
- roll_predicted = F.softmax(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)
-
- 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())
-
- # Total loss
- loss_yaw += alpha * loss_reg_yaw
- loss_pitch += alpha * loss_reg_pitch
- loss_roll += alpha * loss_reg_roll
-
- loss_seq = [loss_yaw, loss_pitch, loss_roll]
- grad_seq = [torch.Tensor(1).cuda(gpu) for _ in range(len(loss_seq))]
- 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_AFLW_finetuned_iter_'+ str(i+1) + '.pkl')
-
- # Save models at numbered epochs.
- if epoch % 1 == 0 and epoch < num_epochs - 1:
- print 'Taking snapshot...'
- torch.save(model.state_dict(),
- 'output/snapshots/resnet50_AFLW_finetuned_epoch_'+ str(epoch+1) + '.pkl')
-
- # Save the final Trained Model
- torch.save(model.state_dict(), 'output/snapshots/resnet50_AFLW_finetuned_epoch_' + str(epoch+1) + '.pkl')
diff --git a/code/train_AFLW_preangles.py b/code/train_AFLW_preangles.py
deleted file mode 100644
index b12149c..0000000
--- a/code/train_AFLW_preangles.py
+++ /dev/null
@@ -1,265 +0,0 @@
-import numpy as np
-import torch
-import torch.nn as nn
-from torch.autograd import Variable
-from torch.utils.data import DataLoader
-from torchvision import transforms
-import torchvision
-import torch.backends.cudnn as cudnn
-import torch.nn.functional as F
-
-import cv2
-import matplotlib.pyplot as plt
-import sys
-import os
-import argparse
-
-import datasets
-import hopenet
-import torch.utils.model_zoo as model_zoo
-
-model_urls = {
- 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth',
- 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth',
- 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth',
- 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth',
- 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth',
-}
-
-def parse_args():
- """Parse input arguments."""
- parser = argparse.ArgumentParser(description='Head pose estimation using the Hopenet network.')
- parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]',
- 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.',
- default=0.001, type=float)
- parser.add_argument('--data_dir', dest='data_dir', help='Directory path for data.',
- 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=0.00, type=float)
- args = parser.parse_args()
- return args
-
-def get_ignored_params(model):
- # Generator function that yields ignored params.
- b = []
- b.append(model.conv1)
- b.append(model.bn1)
- b.append(model.fc_finetune)
- 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 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_fc_params(model):
- b = []
- b.append(model.fc_yaw)
- b.append(model.fc_pitch)
- b.append(model.fc_roll)
- for i in range(len(b)):
- 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
- model_dict = model.state_dict()
- # 1. filter out unnecessary keys
- snapshot = {k: v for k, v in snapshot.items() if k in model_dict}
- # 2. overwrite entries in the existing state dict
- model_dict.update(snapshot)
- # 3. load the new state dict
- model.load_state_dict(model_dict)
-
-if __name__ == '__main__':
- args = parse_args()
-
- cudnn.enabled = True
- num_epochs = args.num_epochs
- num_epochs_ft = args.num_epochs_ft
- batch_size = args.batch_size
- gpu = args.gpu_id
-
- if not os.path.exists('output/snapshots'):
- os.makedirs('output/snapshots')
-
- # 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, 0)
- # 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(),
- transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
-
- pose_dataset = datasets.AFLW(args.data_dir, args.filename_list,
- transformations)
- 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(gpu)
- reg_criterion = nn.MSELoss().cuda(gpu)
- # Regression loss coefficient
- alpha = args.alpha
-
- idx_tensor = [idx for idx in xrange(66)]
- idx_tensor = Variable(torch.FloatTensor(idx_tensor)).cuda(gpu)
-
- 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):
- 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))
-
- 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)
- pitch_predicted = torch.sum(pitch_predicted * idx_tensor, 1)
- roll_predicted = torch.sum(roll_predicted * idx_tensor, 1)
-
- 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())
-
- # print yaw_predicted, label_yaw.float(), loss_reg_yaw
- # Total loss
- loss_yaw += alpha * loss_reg_yaw
- loss_pitch += alpha * loss_reg_pitch
- loss_roll += alpha * loss_reg_roll
-
- loss_seq = [loss_yaw, loss_pitch, loss_roll]
- # loss_seq = [loss_reg_yaw, loss_reg_pitch, loss_reg_roll]
- 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: 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/' + args.output_string + '_iter_'+ str(i+1) + '.pkl')
-
- # Save models at numbered epochs.
- if epoch % 1 == 0 and epoch < num_epochs:
- print 'Taking snapshot...'
- torch.save(model.state_dict(),
- 'output/snapshots/' + args.output_string + '_epoch_'+ str(epoch+1) + '.pkl')
-
- print 'Second phase of training (finetuning layer).'
- for epoch in range(num_epochs_ft):
- for i, (images, 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(labels[:,:3].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.data * idx_tensor, 1)
- pitch_predicted = torch.sum(pitch_predicted.data * idx_tensor, 1)
- roll_predicted = torch.sum(roll_predicted.data * idx_tensor, 1)
-
- 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())
-
- # Total loss
- loss_yaw += alpha * loss_reg_yaw
- loss_pitch += alpha * loss_reg_pitch
- loss_roll += alpha * loss_reg_roll
-
- # Finetuning loss
- loss_angles = reg_criterion(angles[0], label_angles.float())
-
- loss_seq = [loss_yaw, loss_pitch, loss_roll, 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 - 1:
- print 'Taking snapshot...'
- torch.save(model.state_dict(),
- 'output/snapshots/' + args.output_string + '_epoch_'+ str(num_epochs+epoch+1) + '.pkl')
-
-
- # Save the final Trained Model
- torch.save(model.state_dict(), 'output/snapshots/' + args.output_string + '_epoch_' + str(num_epochs+epoch+1) + '.pkl')
diff --git a/code/train_biwi.py b/code/train_biwi.py
deleted file mode 100644
index 6cf7fcd..0000000
--- a/code/train_biwi.py
+++ /dev/null
@@ -1,205 +0,0 @@
-import numpy as np
-import torch
-import torch.nn as nn
-from torch.autograd import Variable
-from torch.utils.data import DataLoader
-from torchvision import transforms
-import torchvision
-import torch.backends.cudnn as cudnn
-import torch.nn.functional as F
-
-import cv2
-import matplotlib.pyplot as plt
-import sys
-import os
-import argparse
-
-import datasets
-import hopenet
-import torch.utils.model_zoo as model_zoo
-
-model_urls = {
- 'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth',
- 'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth',
- 'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth',
- 'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth',
- 'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth',
-}
-
-def parse_args():
- """Parse input arguments."""
- parser = argparse.ArgumentParser(description='Head pose estimation using the Hopenet network.')
- parser.add_argument('--gpu', dest='gpu_id', help='GPU device id to use [0]',
- 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('--batch_size', dest='batch_size', help='Batch size.',
- default=16, type=int)
- parser.add_argument('--lr', dest='lr', help='Base learning rate.',
- default=0.001, type=float)
- parser.add_argument('--data_dir', dest='data_dir', help='Directory path for data.',
- 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('--finetune', dest='finetune', help='Boolean: finetune or from Imagenet pretrain.',
- default=False, type=bool)
- parser.add_argument('--snapshot', dest='snapshot', help='Path to finetune snapshot.',
- default='', type=str)
-
- args = parser.parse_args()
-
- return args
-
-def get_ignored_params(model):
- # Generator function that yields ignored params.
- b = []
- b.append(model.conv1)
- b.append(model.bn1)
- 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
-
-def get_non_ignored_params(model):
- # Generator function that yields params that will be optimized.
- b = []
- b.append(model.fc_yaw)
- b.append(model.fc_pitch)
- b.append(model.fc_roll)
- for i in range(len(b)):
- for j in b[i].modules():
- for k in j.parameters():
- yield k
-
-def load_filtered_state_dict(model, snapshot):
- # By user apaszke from discuss.pytorch.org
- model_dict = model.state_dict()
- # 1. filter out unnecessary keys
- snapshot = {k: v for k, v in snapshot.items() if k in model_dict}
- # 2. overwrite entries in the existing state dict
- model_dict.update(snapshot)
- # 3. load the new state dict
- model.load_state_dict(model_dict)
-
-if __name__ == '__main__':
- args = parse_args()
-
- cudnn.enabled = True
- num_epochs = args.num_epochs
- batch_size = args.batch_size
- gpu = args.gpu_id
-
- if not os.path.exists('output/snapshots'):
- os.makedirs('output/snapshots')
-
- # 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)
- # ResNet18
- # model = hopenet.Hopenet(torchvision.models.resnet.BasicBlock, [2, 2, 2, 2], 66)
-
- if args.finetune:
- print 'Finetuning.'
- model.load_state_dict(torch.load(args.snapshot))
- else:
- 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(224),
- transforms.RandomCrop(224), transforms.ToTensor(),
- transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])
-
- pose_dataset = datasets.BIWI(args.data_dir, args.filename_list,
- transformations)
- train_loader = torch.utils.data.DataLoader(dataset=pose_dataset,
- batch_size=batch_size,
- shuffle=True,
- num_workers=2)
-
- model.cuda(gpu)
- criterion = nn.CrossEntropyLoss().cuda()
- reg_criterion = nn.MSELoss().cuda()
- # Regression loss coefficient
- alpha = 0.01
-
- idx_tensor = [idx for idx in xrange(66)]
- idx_tensor = 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}],
- # lr = args.lr)
-
- optimizer = torch.optim.SGD([{'params': get_ignored_params(model), 'lr': args.lr},
- {'params': get_non_ignored_params(model), 'lr': args.lr}],
- lr = args.lr, momentum=0.9, weight_decay=0.01)
-
- print 'Ready to train network.'
-
- for epoch in range(num_epochs):
- for i, (images, 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))
-
- optimizer.zero_grad()
- model.zero_grad()
-
- yaw, pitch, roll = model(images)
-
- # Cross entropy loss
- loss_yaw = criterion(yaw, label_yaw)
- loss_pitch = criterion(pitch, label_pitch)
- loss_roll = criterion(roll, label_roll)
-
- # MSE loss
- yaw_predicted = F.softmax(yaw)
- pitch_predicted = F.softmax(pitch)
- roll_predicted = F.softmax(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)
-
- 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())
-
- # Total loss
- loss_yaw += alpha * loss_reg_yaw
- loss_pitch += alpha * loss_reg_pitch
- loss_roll += alpha * loss_reg_roll
-
- loss_seq = [loss_yaw, loss_pitch, loss_roll]
- grad_seq = [torch.Tensor(1).cuda(gpu) for _ in range(len(loss_seq))]
- 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_ftbiwi_iter_'+ str(i+1) + '.pkl')
-
- # Save models at numbered epochs.
- if epoch % 1 == 0 and epoch < num_epochs - 1:
- print 'Taking snapshot...'
- torch.save(model.state_dict(),
- 'output/snapshots/resnet50_ftbiwi_epoch_'+ str(epoch+1) + '.pkl')
-
- # Save the final Trained Model
- torch.save(model.state_dict(), 'output/snapshots/resnet50_ftbiwi_epoch_' + str(epoch+1) + '.pkl')
diff --git a/code/train_preangles.py b/code/train_preangles.py
index 5f23b25..6328ef2 100644
--- a/code/train_preangles.py
+++ b/code/train_preangles.py
@@ -197,7 +197,7 @@
loss_pitch += alpha * loss_reg_pitch
loss_roll += alpha * loss_reg_roll
- loss_yaw *= 0.35
+ loss_yaw *= 1
loss_seq = [loss_yaw, loss_pitch, loss_roll]
# loss_seq = [loss_reg_yaw, loss_reg_pitch, loss_reg_roll]
--
Gitblit v1.8.0