def generate_model(opt):
assert opt.model in ['resnet', 'densenet', 'se_resnet']
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(pretrained=True, num_classes=opt.n_classes)
elif opt.model_depth == 18:
model = resnet.resnet18(pretrained=True, num_classes=opt.n_classes)
elif opt.model_depth == 34:
model = resnet.resnet34(pretrained=True, num_classes=opt.n_classes)
elif opt.model_depth == 50:
model = resnet.resnet50(pretrained=True, num_classes=opt.n_classes)
elif opt.model_depth == 101:
model = resnet.resnet101(pretrained=True,
num_classes=opt.n_classes)
elif opt.model_depth == 152:
model = resnet.resnet152(pretrained=True,
num_classes=opt.n_classes)
elif opt.model_depth == 200:
model = resnet.resnet200(pretrained=True,
num_classes=opt.n_classes)
elif opt.model == 'se_resnet':
assert opt.model_depth in [18, 34, 50, 101, 152]
from models.se_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = se_resnet.resnet18(pretrained=True,
num_classes=opt.n_classes)
elif opt.model_depth == 34:
model = se_resnet.resnet34(pretrained=True,
num_classes=opt.n_classes)
elif opt.model_depth == 50:
model = se_resnet.resnet50(pretrained=True,
num_classes=opt.n_classes)
elif opt.model_depth == 101:
model = se_resnet.resnet101(pretrained=True,
num_classes=opt.n_classes)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(config):
model = resnet.resnet101(
num_classes=config.getint('Network', 'classes'),
shortcut_type=config.get('Network', 'resnet_shortcut'),
cardinality=config.getint('Network', 'resnet_cardinality'),
sample_size=config.getint('Network', 'sample_size'),
sample_duration=config.getint('Network', 'sample_duration'),
input_channels=config.getint('Network', 'input_channels'),
)
if config.getboolean('Network', 'use_cuda'):
model = model.cuda()
else:
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
model = nn.DataParallel(model)
pretrain_path = config.get('Network', 'pretrain_path')
if pretrain_path:
print('loading pretrained model {}'.format(pretrain_path))
pretrain = torch.load(pretrain_path)
assert config.get('Network', 'arch') == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
model.module.fc = nn.Linear(
model.module.fc.in_features,
config.getint('Network', 'n_finetune_classes'))
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(
model, config.getint('Network', 'ft_begin_index'))
return model, parameters
return model, model.parameters()
def get_model_param(args):
# assert args.model in ['resnet', 'vgg']
if args.model == 'resnet':
assert args.model_depth in [18, 34, 50, 101, 152]
from models.resnet import get_fine_tuning_parameters
if args.model_depth == 18:
model = resnet.resnet18(pretrained=False,
input_size=args.input_size,
num_classes=args.n_classes)
elif args.model_depth == 34:
model = resnet.resnet34(pretrained=False,
input_size=args.input_size,
num_classes=args.n_classes)
elif args.model_depth == 50:
model = resnet.resnet50(pretrained=False,
input_size=args.input_size,
num_classes=args.n_classes)
elif args.model_depth == 101:
model = resnet.resnet101(pretrained=False,
input_size=args.input_size,
num_classes=args.n_classes)
elif args.model_depth == 152:
model = resnet.resnet152(pretrained=False,
input_size=args.input_size,
num_classes=args.n_classes)
# elif args.model == 'vgg':
# pass
# Load pretrained model here
if args.finetune:
pretrained_model = model_path[args.arch]
args.pretrain_path = os.path.join(args.root_path, 'pretrained_models',
pretrained_model)
print("=> loading pretrained model '{}'...".format(pretrained_model))
model.load_state_dict(torch.load(args.pretrain_path))
# Only modify the last layer
if args.model == 'resnet':
model.fc = nn.Linear(model.fc.in_features, args.n_finetune_classes)
# elif args.model == 'vgg':
# pass
parameters = get_fine_tuning_parameters(model, args.ft_begin_index,
args.lr_mult1, args.lr_mult2)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
model_type=opt.model_type)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
model_type=opt.model_type)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
model_type=opt.model_type)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
model_type=opt.model_type)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
import os
# os.environ['CUDA_VISIBLE_DEVICES'] = f'{opt.cuda_id}'
model = model.cuda(device=opt.cuda_id)
model = nn.DataParallel(model, device_ids=[0]) # CUDA change
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
print(pretrain['arch'])
arch = f'{opt.model}-{opt.model_depth}'
# arch = opt.model + '-' + opt.model_depth
print(arch)
assert arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda(
device=opt.cuda_id)
# elif opt.use_quadriplet:
# model = EmbeddingModel(model, opt.n_finetune_classes, not opt.no_cuda, opt.cuda_id)
else:
model.module.fc = nn.Sequential(
nn.Dropout(0.4),
nn.Linear(model.module.fc.in_features, 512), nn.ReLU6(),
nn.Dropout(0.4), nn.Linear(512, 128), nn.ReLU6(),
nn.Linear(128,
opt.n_finetune_classes)).cuda(device=opt.cuda_id)
# model.module.fc = nn.Linear(model.module.fc.in_features,
# opt.n_finetune_classes)
# model.module.fc = model.module.fc.cuda(device=opt.cuda_id)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
print(len(list(parameters)), 'params to fine tune')
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'resnext_fa', 'densenet', 'p3d'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'p3d':
assert opt.model_depth in [50, 101, 152]
if opt.model_depth == 50:
model = p3d.P3D63(num_classes=opt.n_classes)
elif opt.model_depth == 101:
model = p3d.P3D131(num_classes=opt.n_classes)
elif opt.model_depth == 152:
model = p3d.P3D199(num_classes=opt.n_classes)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext_fa import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext_fa':
assert opt.model_depth in [50, 101, 152]
from models.resnext_fa import get_fine_tuning_parameters, get_fine_tuning_parameters_fa
if opt.model_depth == 50:
model = resnext_fa.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext_fa.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext_fa.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
#pdb.set_trace();
#assert opt.arch == pretrain['arch']
model_dict = model.state_dict();
#pdb.set_trace();
model_dict.update(pretrain['state_dict']);
model.load_state_dict(model_dict);
#model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
# do not need to add new fc layer when finetuning model has the same class num
elif (opt.n_classes != opt.n_finetune_classes):
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
if (opt.model == 'resnext_fa'):
parameters = get_fine_tuning_parameters_fa(model, opt.learning_rate)
else:
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in ['resnet']
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
input_chan = 3
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, \
input_chan=input_chan)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, \
input_chan=input_chan)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, \
input_chan=input_chan)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, \
input_chan=input_chan)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
if 'arch' in pretrain:
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
else:
if "state_dict" in pretrain.keys():
model.module.load_state_dict(pretrain['state_dict'])
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
128)
model.load_state_dict(pretrain['model_state_dict'])
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
model.module.freeze_layers(opt.ft_begin_index)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
model.fc = nn.Linear(model.fc.in_features, opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
###################################################################
# ResNet
###################################################################
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
###################################################################
# Wider ResNet
###################################################################
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
###################################################################
# ResNext
###################################################################
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
###################################################################
# Pre-ResNet
###################################################################
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
###################################################################
# DenseNet
###################################################################
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
###################################################################
# Finalizing the model
###################################################################
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=opt.device_ids)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch'] # ensure that pretrain model is the same architecture
model.load_state_dict(pretrain['state_dict'])
# change the fc layer output size
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
#
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'resnet_skeleton', 'preresnet', 'wideresnet', 'resnext',
'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet_skeleton':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet_skeleton import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet_skeleton.resnet_skeleton10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet_skeleton.resnet_skeleton18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet_skeleton.resnet_skeleton34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet_skeleton.resnet_skeleton50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet_skeleton.resnet_skeleton101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet_skeleton.resnet_skeleton152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet_skeleton.resnet_skeleton200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
if opt.cuda_id is None:
model = model.cuda()
else:
model = model.cuda(opt.cuda_id)
# model = nn.DataParallel(model, device_ids=None)
if opt.cuda_id is None:
model = nn.DataParallel(model, device_ids=None)
else:
model = nn.DataParallel(model, device_ids=[opt.cuda_id])
if opt.pretrain_path:
print(' loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
if opt.model == 'resnet_skeleton':
pretrained_dict = pretrain['state_dict']
model_dict = model.state_dict()
# print('----------------')
# for k, v in pretrained_dict.items():
# if k in model_dict:
# print(k)
# print('----------------')
# pretrained_dict = {k: v for k, v in pretrained_dict.items() if k in model_dict}
pretrained_dict = {
k: v
for k, v in pretrained_dict.items()
if k in model_dict and 'fc' not in k
} ## for concatenate
model_dict.update(pretrained_dict)
model.load_state_dict(model_dict)
else:
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
if opt.cuda_id is None:
model.module.classifier = model.module.classifier.cuda()
else:
model.module.classifier = model.module.classifier.cuda(
opt.cuda_id)
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
if opt.cuda_id is None:
model.module.fc = model.module.fc.cuda()
else:
model.module.fc = model.module.fc.cuda(opt.cuda_id)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
# import pdb;pdb.set_trace()
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet', 'se_resnet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model == 'se_resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.se_resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = se_resnet.se_resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 18:
model = se_resnet.se_resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 34:
model = se_resnet.se_resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 50:
model = se_resnet.se_resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 101:
model = se_resnet.se_resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 152:
model = se_resnet.se_resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 200:
model = se_resnet.se_resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
channels=opt.channels)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
pretrain_dict = pretrain['state_dict']
model_dict = model.state_dict()
pretrain_dict = {
k: v
for k, v in pretrain_dict.items() if k in model_dict
}
# 更新现有的model_dict
w = pretrain_dict['module.conv1.weight']
pretrain_dict['module.conv1.weight'] = torch.nn.Parameter(
w[:, :1, :, :])
w_fc = pretrain_dict['module.fc.weight']
pretrain_dict['module.fc.weight'] = torch.nn.Parameter(
w_fc[:opt.n_finetune_classes, :])
w_bias = pretrain_dict['module.fc.bias']
pretrain_dict['module.fc.bias'] = torch.nn.Parameter(
w_bias[:opt.n_finetune_classes])
model_dict.update(pretrain_dict)
model.load_state_dict(model_dict)
#model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet','standard'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'standard':
from models.C3D_model import get_fine_tuning_parameters
num_classes=opt.n_classes
model = C3D_model.C3D(num_classes)
#s1m = torch.load('c3d.pickle')
#reset last layer to 400 classes
#s1m['fc8.weight']= torch.FloatTensor(num_classes,4096)
#reset bias to tensor of size 400
#s1m['fc8.bias']= torch.FloatTensor(num_classes)
#load weights into C3D model
#model.load_state_dict(s1m)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
elif opt.model =='resnet':
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
else:
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
elif opt.model == 'resnet':
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
else:
model.module.fc8 = nn.Linear(model.module.fc8.in_features, opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt, modality):
assert opt.model in ['resnet', 'resnetl', 'resnext', 'c3d']
if opt.model == 'resnet':
# assert opt.model_depth in [10]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.n_frames_per_clip)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.n_frames_per_clip)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.n_frames_per_clip)
elif opt.model == 'resnetl':
assert opt.model_depth in [10]
from models.resnetl import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnetl.resnetl10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.n_frames_per_clip)
elif opt.model == 'resnext':
assert opt.model_depth in [101]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.n_frames_per_clip)
elif opt.model == 'c3d':
assert opt.model_depth in [10]
from models.c3d import get_fine_tuning_parameters
if opt.model_depth == 10:
model = c3d.c3d_v1(sample_size=opt.sample_size,
sample_duration=opt.n_frames_per_clip,
num_classes=opt.n_classes)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path, map_location='cpu')
# assert opt.arch == pretrain['arch']
feature_state_dict = {
key: value
for key, value in pretrain['state_dict'].items()
if key not in ['module.fc.weight', 'module.fc.bias']
}
state_dict = deepcopy(model.state_dict())
state_dict.update(feature_state_dict)
model.load_state_dict(state_dict)
# model.load_state_dict(pretrain['state_dict'])
if modality == 'RGB' and opt.model != 'c3d':
print("[INFO]: RGB model is used for init model")
# model = _modify_first_conv_layer(model,3,3) ##### Check models trained (3,7,7) or (7,7,7)
elif modality == 'Depth':
print(
"[INFO]: Converting the pretrained model to Depth init model")
model = _construct_depth_model(model)
print("[INFO]: Done. Flow model ready.")
elif modality == 'RGB-D':
print(
"[INFO]: Converting the pretrained model to RGB+D init model")
model = _construct_rgbdepth_model(model)
print("[INFO]: Done. RGB-D model ready.")
# Check first kernel size
modules = list(model.modules())
first_conv_idx = list(
filter(lambda x: isinstance(modules[x], nn.Conv3d),
list(range(len(modules)))))[0]
conv_layer = modules[first_conv_idx]
if conv_layer.kernel_size[0] > opt.n_frames_per_clip:
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(model,
int(opt.n_frames_per_clip / 2), 1)
if opt.model == 'c3d': # CHECK HERE
model.module.fc = nn.Linear(model.module.fc[0].in_features,
model.module.fc[0].out_features)
model.module.fc = model.module.fc.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path, map_location='cpu')
# assert opt.arch == pretrain['arch']
feature_state_dict = {
key.replace('module.', ''): value
for key, value in pretrain['state_dict'].items()
if key not in ['module.fc.weight', 'module.fc.bias']
}
state_dict = deepcopy(model.state_dict())
state_dict.update(feature_state_dict)
model.load_state_dict(state_dict)
# model.load_state_dict(pretrain['state_dict'])
if modality == 'RGB' and opt.model != 'c3d':
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(model, 3, 3)
elif modality == 'Depth':
print(
"[INFO]: Converting the pretrained model to Depth init model")
model = _construct_depth_model(model)
print("[INFO]: Deoth model ready.")
elif modality == 'RGB-D':
print(
"[INFO]: Converting the pretrained model to RGB-D init model")
model = _construct_rgbdepth_model(model)
print("[INFO]: Done. RGB-D model ready.")
# Check first kernel size
modules = list(model.modules())
first_conv_idx = list(
filter(lambda x: isinstance(modules[x], nn.Conv3d),
list(range(len(modules)))))[0]
conv_layer = modules[first_conv_idx]
if conv_layer.kernel_size[0] > opt.n_frames_per_clip:
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(model,
int(opt.n_frames_per_clip / 2), 1)
if opt.model == 'c3d': # CHECK HERE
model.fc = nn.Linear(model.fc[0].in_features,
model.fc[0].out_features)
else:
model.fc = nn.Linear(model.fc.in_features, opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
elif opt.model_depth == 18:
model = resnet.resnet18(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
elif opt.model_depth == 34:
model = resnet.resnet34(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
elif opt.model_depth == 50:
model = resnet.resnet50(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
elif opt.model_depth == 101:
model = resnet.resnet101(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
elif opt.model_depth == 152:
model = resnet.resnet152(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
elif opt.model_depth == 200:
model = resnet.resnet200(shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
audiovisual=opt.audiovisual)
else:
raise ValueError('Model type not supported')
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'], strict=0)
if opt.audio_pretrain_path:
print('loading pretrained audio model {}'.format(
opt.audio_pretrain_path))
pretrain_audio = torch.load(opt.audio_pretrain_path)
model.load_state_dict(pretrain_audio, strict=0)
(parameters,
name_parameters) = get_fine_tuning_parameters(model,
opt.learning_rate_sal,
opt.weight_decay)
return model, parameters
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
#assert opt.arch == pretrain['arch']
pretrained_dict = pretrain['state_dict']
# print("PRETRAIN BEFORE:", pretrained_dict.keys())
model_dict = model.state_dict()
#print("Current model:", model.state_dict().keys())
#print("Pretrained model:", pretrain['state_dict'].keys())
# 1. filter out unnecessary keys
#pretrained_dict = {k: v for k, v in pretrained_dict.items() if k in model_dict}
# 2. overwrite entries in the existing state dict
pretrained_dict.update(model_dict)
#print("Pretrained model after:", len(model_dict.keys()))
# 3. load the new state dict
model.load_state_dict(pretrained_dict)
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
#assert opt.arch == pretrain['arch']
print("pretrain", pretrain['state_dict'])
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if(opt.pretrain_path):
if(opt.is_rgb):
ch = 3
elif(opt.is_depth):
ch = 1
elif(opt.is_rgb and opt.is_depth):
ch = 4
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration, channel=ch)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model_dict = model.state_dict()
pretrained_dict = pretrain['state_dict']
# 1. filter out unnecessary keys
if((not opt.is_rgb and opt.is_depth) or (opt.is_rgb and opt.is_depth) ):
pretrained_dict = {k: v for k, v in pretrained_dict.items() if k != 'module.conv1.weight'}
#for k, v in pretrained_dict.items():
# print(k)
# 2. overwrite entries in the existing state dict
model_dict.update(pretrained_dict)
# 3. load the new state dict
model.load_state_dict(model_dict)
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model
def generate_model(opt):
assert opt.clip_model in ['resnet', 'r2plus1d', 'avts', 'i3d']
if opt.clip_model == 'resnet':
assert opt.clip_model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
from models.resnet import get_fine_tuning_parameters_layer_lr
if opt.clip_model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.clip_model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.clip_model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.clip_model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
no_last_fc=opt.no_last_fc)
elif opt.clip_model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.clip_model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.clip_model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.clip_model.lower() in ['r2+1d', 'r2.5d', 'r2plus1d']:
from models.r2plus1d import get_fine_tuning_parameters, get_fine_tuning_parameters_layer_lr
print("Making R2+1D model, depth", opt.clip_model_depth)
model = r2plus1d.r2plus1d_34(num_classes=opt.n_classes)
elif opt.clip_model.lower() in ['avts']:
from models.avts import get_fine_tuning_parameters, get_fine_tuning_parameters_layer_lr
print("Making AVTS model")
model = avts.mc3_avts()
model.add_module("flatten", torch.nn.Flatten(1))
model.add_module("fc", nn.Linear(256, opt.n_classes))
elif opt.clip_model == 'i3d':
from models import i3res
if opt.clip_model_depth == 50:
print("Making I3D model, depth", opt.clip_model_depth)
model = i3res.i3res_50(sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
num_classes=opt.n_classes)
elif opt.clip_model_depth == 34:
print("Making I3D model, depth", opt.clip_model_depth)
model = i3res.i3res_34(sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
num_classes=opt.n_classes)
if opt.distributed:
# For multiprocessing distributed, DistributedDataParallel constructor
# should always set the single device scope, otherwise,
# DistributedDataParallel will use all available devices.
if opt.gpu is not None:
torch.cuda.set_device(opt.gpu)
model.cuda(opt.gpu)
# When using a single GPU per process and per
# DistributedDataParallel, we need to divide the batch size
# ourselves based on the total number of GPUs we have
opt.batch_size = int(opt.batch_size / opt.ngpus_per_node)
opt.n_threads = int(opt.n_threads / opt.ngpus_per_node)
model = torch.nn.parallel.DistributedDataParallel(
model, device_ids=[opt.gpu])
else:
model.cuda()
# DistributedDataParallel will divide and allocate batch_size to all
# available GPUs if device_ids are not set
model = torch.nn.parallel.DistributedDataParallel(model)
elif opt.gpu is not None:
torch.cuda.set_device(opt.gpu)
model = model.cuda(opt.gpu)
else:
# DataParallel will divide and allocate batch_size to all available GPUs
model = torch.nn.DataParallel(model).cuda()
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
model.load_state_dict(pretrain['state_dict'])
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
if opt.layer_lr is not None:
parameters = get_fine_tuning_parameters_layer_lr(
model, opt.ft_begin_index, opt.layer_lr)
else:
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
# else:
# if opt.layer_lr is not None:
# parameters = get_fine_tuning_parameters_layer_lr(model, opt.ft_begin_index, opt.layer_lr)
# else:
# parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
# return model, parameters
return model
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
# strip off the 'module.' for each module; this get's added when a model is saved using nn.DataParallel
pretrain['state_dict'] = {
k[7:]: v
for k, v in pretrain['state_dict'].items()
}
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet',
'preresnet',
'wideresnet',
'resnext',
'densenet',
'c3d',
'c2d',
'c2d_exp',
'c2d_coord',
'c3d_color',
'c2d_pt',
'c2d_pt2',
'c2d_pt5',
'c2d_pt7',
'c2d_pt_exp',
'c2d_pt2_exp',
'c2d_pt5_exp',
'c2d_pt_exp_avg',
'c2d_pt_exp_sep',
'c3d_pt_exp',
'c2d_pt_exp_init',
'c2d_pt_expc',
'resnet18_exp',
'resnet34_exp',
'resnet50_exp',
'resnet101_exp',
'resnet152_exp',
'resnext50_32x4d_exp',
'resnext101_32x8d_exp',
'wide_resnet50_2_exp',
'wide_resnet101_2_exp',
'resnet18_pt_exp',
'resnet34_pt_exp',
'resnet50_pt_exp',
'resnet101_pt_exp',
'resnet152_pt_exp',
'resnext50_32x4d_pt_exp',
'resnext101_32x8d_pt_exp',
'wide_resnet50_2_pt_exp',
'wide_resnet101_2_pt_exp',
# decoder
'stsrresnetexp',
'spc',
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c3d':
model = c3d.C3D(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c3d_color':
model = c3d_color.C3D(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'spc':
model = spc.SPC(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d':
model = c2d.C2D(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt':
model = c2d_pt.C2DPt(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt2':
model = c2d_pt2.C2DPt(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt5':
model = c2d_pt5.C2DPt(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt7':
model = c2d_pt7.C2DPt(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_exp':
model = c2d_exp.C2DExp(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt_exp':
model = c2d_pt_exp.C2DPtExp(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt_expc':
model = c2d_pt_expc.C2DPtExpC(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt_exp_init':
model = c2d_pt_exp_init.C2DPtExp(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c3d_pt_exp':
model = c3d_pt_exp.C3DPtExp(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt_exp_avg':
model = c2d_pt_exp_avg.C2DPtExpAvg(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt_exp_sep':
model = c2d_pt_exp_sep.C2DPtExpSep(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt5_exp':
model = c2d_pt5_exp.C2DPtExp(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_pt2_exp':
model = c2d_pt2_exp.C2DPtExp(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c2d_coord':
model = c2d_coord.C2DCoord(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet18_exp':
model = resnet_exp.resnet18(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet34_exp':
model = resnet_exp.resnet34(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet50_exp':
model = resnet_exp.resnet50(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet101_exp':
model = resnet_exp.resnet101(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet152_exp':
model = resnet_exp.resnet152(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext50_32x4d_exp':
model = resnet_exp.resnext50_32x4d(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext101_32x8d_exp':
model = resnet_exp.resnext101_32x8d(
pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'wide_resnet50_2_exp':
model = resnet_exp.wide_resnet50_2(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'wide_resnet101_2_exp':
model = resnet_exp.wide_resnet101_2(
pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet18_pt_exp':
model = resnet_pt_exp.resnet18(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet34_pt_exp':
model = resnet_pt_exp.resnet34(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet50_pt_exp':
model = resnet_pt_exp.resnet50(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet101_pt_exp':
model = resnet_pt_exp.resnet101(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet152_pt_exp':
model = resnet_pt_exp.resnet152(pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext50_32x4d_pt_exp':
model = resnet_pt_exp.resnext50_32x4d(
pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext101_32x8d_pt_exp':
model = resnet_pt_exp.resnext101_32x8d(
pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'wide_resnet50_2_pt_exp':
model = resnet_pt_exp.wide_resnet50_2(
pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'wide_resnet101_2_pt_exp':
model = resnet_pt_exp.wide_resnet101_2(
pretrained=False,
progress=True,
num_classes=opt.n_classes,
sample_duration=opt.sample_duration)
elif opt.model == 'stsrresnetexp':
model = decoder.STSRResNetExp(sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
args = {
"num_classes": opt.n_classes,
"shortcut_type": opt.resnet_shortcut,
"sample_size": opt.sample_size,
"sample_duration": opt.sample_duration
}
if opt.model_depth == 10:
model = resnet.resnet10(**args)
elif opt.model_depth == 18:
model = resnet.resnet18(**args)
elif opt.model_depth == 34:
model = resnet.resnet34(**args)
elif opt.model_depth == 50:
model = resnet.resnet50(**args)
elif opt.model_depth == 101:
model = resnet.resnet101(**args)
elif opt.model_depth == 152:
model = resnet.resnet152(**args)
elif opt.model_depth == 200:
model = resnet.resnet200(**args)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
args = {
"num_classes": opt.n_classes,
"shortcut_type": opt.resnet_shortcut,
"cardinality": opt.resnext_cardinality,
"sample_size": opt.sample_size,
"sample_duration": opt.sample_duration
}
if opt.model_depth == 50:
model = resnext.resnet50(**args)
elif opt.model_depth == 101:
model = resnext.resnet101(**args)
elif opt.model_depth == 152:
model = resnext.resnet152(**args)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
args = {
"num_classes": opt.n_classes,
"shortcut_type": opt.resnet_shortcut,
"sample_size": opt.sample_size,
"sample_duration": opt.sample_duration
}
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(**args)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(**args)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(**args)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(**args)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(**args)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(**args)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
args = {
"num_classes": opt.n_classes,
"sample_size": opt.sample_size,
"sample_duration": opt.sample_duration
}
if opt.model_depth == 121:
model = densenet.densenet121(**args)
elif opt.model_depth == 169:
model = densenet.densenet169(**args)
elif opt.model_depth == 201:
model = densenet.densenet201(**args)
elif opt.model_depth == 264:
model = densenet.densenet264(**args)
if opt.no_cuda:
device = 'cpu'
else:
device = 'cuda'
model = model.to(device)
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path, map_location=device)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.to(device)
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.to(device)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet3D', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet3D':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
# model = nn.DataParallel(model, device_ids=None)
if opt.pretrain:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain:
print('loading pretrained model {}'.format(opt.pretrain))
pretrain = torch.load(opt.pretrain)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
# import torch.nn as nn
# import math
# import numpy as np
# import torch
# from torch.autograd import Variable
# from torch.nn.parameter import Parameter
# import random
# import torch.utils.model_zoo as model_zoo
#
# def initweights(m):
# orthogonal_flag = False
# for layer in m.modules():
# if isinstance(layer, nn.Conv2d):
# n = layer.kernel_size[0] * layer.kernel_size[1] * layer.out_channels
# layer.weight.data.normal_(0, math.sqrt(2. / n))
#
#
#
# # orthogonal initialize
# """Reference:
# [1] Saxe, Andrew M., James L. McClelland, and Surya Ganguli.
# "Exact solutions to the nonlinear dynamics of learning in deep
# linear neural networks." arXiv preprint arXiv:1312.6120 (2013)."""
# if orthogonal_flag:
# weight_shape = layer.weight.data.cpu().numpy().shape
# u, _, v = np.linalg.svd(layer.weight.data.cpu().numpy(), full_matrices=False)
# flat_shape = (weight_shape[0], np.prod(weight_shape[1:]))
# q = u if u.shape == flat_shape else v
# q = q.reshape(weight_shape)
# layer.weight.data.copy_(torch.Tensor(q))
#
# elif isinstance(layer, nn.BatchNorm2d):
# layer.weight.data.fill_(1)
# layer.bias.data.zero_()
# elif isinstance(layer, nn.Linear):
# layer.bias.data.zero_()
#
# def conv3x3(in_planes, out_planes, stride=1):
# """3x3 convolution with padding"""
# return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
# padding=1, bias=False)
#
# class BasicBlock(nn.Module):
# expansion = 1
#
# def __init__(self, inplanes, planes, stride=1, downsample=None):
# super(BasicBlock, self).__init__()
# self.conv1 = conv3x3(inplanes, planes, stride)
# self.bn1 = nn.BatchNorm2d(planes)
# self.relu = nn.ReLU(inplace=True)
# self.conv2 = conv3x3(planes, planes)
# self.bn2 = nn.BatchNorm2d(planes)
# self.downsample = downsample
# self.stride = stride
#
# def forward(self, x):
# residual = x
#
# out = self.conv1(x)
# out = self.bn1(out)
# out = self.relu(out)
#
# out = self.conv2(out)
# out = self.bn2(out)
#
# if self.downsample is not None:
# residual = self.downsample(x)
#
# out += residual
# out = self.relu(out)
#
# return out
#
# class Bottleneck(nn.Module):
# expansion = 4
#
# def __init__(self, inplanes, planes, stride=1, downsample=None):
# super(Bottleneck, self).__init__()
# self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
# self.bn1 = nn.BatchNorm2d(planes)
# self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
# padding=1, bias=False)
# self.bn2 = nn.BatchNorm2d(planes)
# self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
# self.bn3 = nn.BatchNorm2d(planes * 4)
# self.relu = nn.ReLU(inplace=True)
# self.downsample = downsample
# self.stride = stride
#
# def forward(self, x):
# residual = x
#
# out = self.conv1(x)
# out = self.bn1(out)
# out = self.relu(out)
#
# out = self.conv2(out)
# out = self.bn2(out)
# out = self.relu(out)
#
# out = self.conv3(out)
# out = self.bn3(out)
#
# if self.downsample is not None:
# residual = self.downsample(x)
#
# out += residual
# out = self.relu(out)
#
# return out
#
#
# class ResNet(nn.Module):
#
# def __init__(self, block, layers, num_classes=1000):
# self.inplanes = 64
# super(ResNet, self).__init__()
# self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
# bias=False)
# self.bn1 = nn.BatchNorm2d(64)
# self.relu = nn.ReLU(inplace=True)
# self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
# self.layer1 = self._make_layer(block, 64, layers[0])
# self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
# self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
# self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
# self.avgpool = nn.AvgPool2d(7, stride=1)
# self.fc = nn.Linear(512 * block.expansion, num_classes)
#
# self.apply(initweights)
#
# def _make_layer(self, block, planes, blocks, stride=1):
# downsample = None
# if stride != 1 or self.inplanes != planes * block.expansion:
# downsample = nn.Sequential(
# nn.Conv2d(self.inplanes, planes * block.expansion,
# kernel_size=1, stride=stride, bias=False),
# nn.BatchNorm2d(planes * block.expansion),
# )
#
# layers = []
# layers.append(block(self.inplanes, planes, stride, downsample))
# self.inplanes = planes * block.expansion
# for i in range(1, blocks):
# layers.append(block(self.inplanes, planes))
#
# return nn.Sequential(*layers)
#
# def forward(self, x):
# x = self.conv1(x)
# x = self.bn1(x)
# x = self.relu(x)
# x = self.maxpool(x)
#
# x = self.layer1(x)
# x = self.layer2(x)
# x = self.layer3(x)
# x = self.layer4(x)
#
# x = self.avgpool(x)
# x = x.view(x.size(0), -1)
# x = self.fc(x)
#
# return x
#
# class ResNetImageNet(nn.Module):
#
# def __init__(self, opt, num_classes=1000):
# super(ResNetImageNet, self).__init__()
# self.opt = opt
# self.depth = opt.depth
# self.num_classes = num_classes
# if self.depth == 18:
# self.model = ResNet(BasicBlock, [2, 2, 2, 2], self.num_classes)
# elif self.depth == 34:
# self.model = ResNet(BasicBlock, [3, 4, 6, 3], self.num_classes)
# elif self.depth == 50:
# self.model = ResNet(Bottleneck, [3, 4, 6, 3], self.num_classes)
# elif self.depth == 101:
# self.model = ResNet(Bottleneck, [3, 4, 23, 3], self.num_classes)
# elif self.depth == 152:
# self.model = ResNet(Bottleneck, [3, 8, 36, 3], self.num_classes)
#
#
# def forward(self, x):
# x = self.model(x)
# return x
#
#
#
# # def resnet18(pretrained=False, **kwargs):
# # """Constructs a ResNet-18 model.
# # Args:
# # pretrained (bool): If True, returns a model pre-trained on ImageNet
# # """
# # model = ResNet(BasicBlock, [2, 2, 2, 2], **kwargs)
# # if pretrained:
# # model.load_state_dict(model_zoo.load_url(model_urls['resnet18']))
# # return model
def generate_model(opt):
assert opt.model in [
'resnet', 'resnet_AE', 'resnet_mask', 'resnet_comp', 'unet', 'icnet',
'icnet_res', 'icnet_res_2D', 'icnet_res_2Dt', 'icnet_DBI',
'icnet_deep', 'icnet_deep_gate', 'icnet_deep_gate_2step'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnet_AE':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet_AE import get_fine_tuning_parameters
if opt.model_depth == 18:
model = resnet_AE.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model_depth == 34:
model = resnet_AE.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model_depth == 50:
model = resnet_AE.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model == 'resnet_mask':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet_mask import get_fine_tuning_parameters
if opt.model_depth == 18:
model = resnet_mask.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model_depth == 34:
model = resnet_mask.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model_depth == 50:
model = resnet_mask.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model == 'resnet_comp':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet_comp import get_fine_tuning_parameters
if opt.model_depth == 18:
model = resnet_comp.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model_depth == 34:
model = resnet_comp.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model_depth == 50:
model = resnet_comp.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
is_gray=opt.is_gray,
opt=opt)
elif opt.model == 'unet':
model = unet_mask.UNet3D(opt=opt)
elif opt.model == 'icnet':
model = icnet_mask.ICNet3D(opt=opt)
elif opt.model == 'icnet_res':
model = icnet_res.ICNetResidual3D(opt=opt)
elif opt.model == 'icnet_res_2D':
model = icnet_res.ICNetResidual2D(opt=opt)
elif opt.model == 'icnet_res_2Dt':
model = icnet_res.ICNetResidual2Dt(opt=opt)
elif opt.model == 'icnet_DBI':
model = icnet_res.ICNetResidual_DBI(opt=opt)
elif opt.model == 'icnet_deep':
model = icnet_res.ICNetDeep(opt=opt)
elif opt.model == 'icnet_deep_gate':
model = icnet_res.ICNetDeepGate(opt=opt)
elif opt.model == 'icnet_deep_gate_2step':
model = icnet_res.ICNetDeepGate2step(opt=opt)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
print('loading from', pretrain['arch'])
child_dict = model.state_dict()
if opt.two_step and opt.test:
parent_list = pretrain['state_dict_1'].keys()
else:
parent_list = pretrain['state_dict'].keys()
print('Not loaded :')
parent_dict = {}
for chi, _ in child_dict.items():
# pdb.set_trace()
# if ('coarse' in chi):
# chi_ori = chi
# chi = 'module.' + ".".join(chi_ori.split('.')[2:])
if chi in parent_list:
if opt.two_step and opt.test:
parent_dict[chi] = pretrain['state_dict_1'][chi]
else:
parent_dict[chi] = pretrain['state_dict'][chi]
else:
print(chi)
print('length :', len(parent_dict.keys()))
child_dict.update(parent_dict)
model.load_state_dict(child_dict)
if not opt.is_AE:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
return model, model.parameters()
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert pretrain['arch'] in ['resnet', 'resnet_AE']
model.load_state_dict(pretrain['state_dict'])
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
if not opt.no_cuda:
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'resnetl', 'resnext', 'c3d', 'mobilenetv2', 'shufflenetv2',
"mstcn"
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 50]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnetl':
assert opt.model_depth in [10, 18]
from models.resnetl import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnetl.resnetl10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnetl.resnetl10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [101]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c3d':
assert opt.model_depth in [10]
from models.c3d import get_fine_tuning_parameters
if opt.model_depth == 10:
model = c3d.c3d_v1(sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
num_classes=opt.n_classes)
elif opt.model == 'mobilenetv2':
from models.mobilenetv2 import get_fine_tuning_parameters
model = mobilenetv2.mob_v2(num_classes=opt.n_classes,
sample_size=opt.sample_size,
width_mult=opt.width_mult)
elif opt.model == 'shufflenetv2':
from models.shufflenetv2 import get_fine_tuning_parameters
model = shufflenetv2.shf_v2(num_classes=opt.n_classes,
sample_size=opt.sample_size,
width_mult=opt.width_mult)
elif opt.model == "mstcn":
model = MultiStageTemporalConvNet(
embed_size=opt.embedding_dim,
encoder=opt.tcn_encoder,
n_classes=opt.n_classes,
input_size=(opt.sample_size, opt.sample_size),
input_channels=4 if opt.modality != "RGB" else 3,
num_stages=opt.tcn_stages,
causal_config=opt.tcn_causality,
CTHW_layout=True,
use_preprocessing=opt.use_preprocessing)
if not opt.no_cuda:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
if opt.pretrain_dataset == 'jester':
if opt.sample_duration < 32 and opt.model != 'c3d':
model = _modify_first_conv_layer(model, 3, 3)
if opt.model in ['mobilenetv2', 'shufflenetv2']:
del pretrain['state_dict']['module.classifier.1.weight']
del pretrain['state_dict']['module.classifier.1.bias']
else:
del pretrain['state_dict']['module.fc.weight']
del pretrain['state_dict']['module.fc.bias']
model.load_state_dict(pretrain['state_dict'], strict=False)
if opt.modality in ['RGB', 'flo'] and opt.model != 'c3d':
print("[INFO]: RGB model is used for init model")
if opt.dataset != 'jester' and not opt.no_first_lay:
model = _modify_first_conv_layer(
model, 3, 3) ##### Check models trained (3,7,7) or (7,7,7)
elif opt.modality in ['Depth', 'seg']:
print(
"[INFO]: Converting the pretrained model to Depth init model")
model = _construct_depth_model(model)
print("[INFO]: Done. Flow model ready.")
elif opt.modality in ['RGB-D', 'RGB-flo', 'RGB-seg']:
if opt.model != "mstcn":
print(
"[INFO]: Converting the pretrained model to RGB+D init model"
)
model = _construct_rgbdepth_model(model)
if opt.no_first_lay:
model = _modify_first_conv_layer(
model, 3,
4) ##### Check models trained (3,7,7) or (7,7,7)
print("[INFO]: Done. RGB-D model ready.")
if opt.pretrain_dataset == opt.dataset:
model.load_state_dict(pretrain['state_dict'])
elif opt.pretrain_dataset in ['egogesture', 'nv', 'denso']:
del pretrain['state_dict']['module.fc.weight']
del pretrain['state_dict']['module.fc.bias']
model.load_state_dict(pretrain['state_dict'], strict=False)
# Check first kernel size
if opt.model != "mstcn":
modules = list(model.modules())
first_conv_idx = list(
filter(lambda x: isinstance(modules[x], nn.Conv3d),
list(range(len(modules)))))[0]
conv_layer = modules[first_conv_idx]
if conv_layer.kernel_size[0] > opt.sample_duration:
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(model,
int(opt.sample_duration / 2),
1)
if opt.model == 'c3d': # CHECK HERE
model.module.fc = nn.Linear(model.module.fc[0].in_features,
model.module.fc[0].out_features)
model.module.fc = model.module.fc.cuda()
elif opt.model in ['mobilenetv2', 'shufflenetv2']:
model.module.classifier = nn.Sequential(
nn.Dropout(0.9),
nn.Linear(model.module.classifier[1].in_features,
opt.n_finetune_classes))
model.module.classifier = model.module.classifier.cuda()
elif opt.model != "mstcn":
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
if opt.model != "mstcn":
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
else:
parameters = model.trainable_parameters()
model = nn.DataParallel(model, device_ids=None)
model = model.cuda()
return model, parameters
else:
print('ERROR no cuda')
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet',
'mobilenet', 'mobilenetv2'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'mobilenet':
from models.mobilenet import get_fine_tuning_parameters
model = mobilenet.get_model(num_classes=opt.n_classes,
sample_size=opt.sample_size,
width_mult=opt.width_mult)
elif opt.model == 'mobilenetv2':
from models.mobilenetv2 import get_fine_tuning_parameters
model = mobilenetv2.get_model(num_classes=opt.n_classes,
sample_size=opt.sample_size,
width_mult=opt.width_mult)
if not opt.no_cuda:
if not opt.no_cuda_predict:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
print("Pretrain arch", pretrain['arch'])
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
ft_begin_index = opt.ft_begin_index
if opt.model in [
'mobilenet', 'mobilenetv2', 'shufflenet', 'shufflenetv2'
]:
model.module.classifier = nn.Sequential(
nn.Dropout(0.9),
nn.Linear(model.module.classifier[1].in_features,
opt.n_finetune_classes))
model.module.classifier = model.module.classifier.cuda()
ft_begin_index = 'complete' if ft_begin_index == 0 else 'last_layer'
elif opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
print("Finetuning at:", ft_begin_index)
parameters = get_fine_tuning_parameters(model, ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
ft_begin_index = opt.ft_begin_index
if opt.model in [
'mobilenet', 'mobilenetv2', 'shufflenet', 'shufflenetv2'
]:
model.module.classifier = nn.Sequential(
nn.Dropout(0.9),
nn.Linear(model.module.classifier[1].in_features,
opt.n_finetune_classes))
model.module.classifier = model.module.classifier.cuda()
ft_begin_index = 'complete' if ft_begin_index == 0 else 'last_layer'
elif opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
print("Finetuning at:", ft_begin_index)
parameters = get_fine_tuning_parameters(model, ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
# if opt.model == 'resnet':
# assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
# from models.resnet import get_fine_tuning_parameters
# if opt.model_depth == 10:
# model = resnet.resnet10(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 18:
# model = resnet.resnet18(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 34:
# model = resnet.resnet34(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 50:
# model = resnet.resnet50(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 101:
# model = resnet.resnet101(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 152:
# model = resnet.resnet152(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 200:
# model = resnet.resnet200(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model == 'wideresnet':
# assert opt.model_depth in [50]
# from models.wide_resnet import get_fine_tuning_parameters
# if opt.model_depth == 50:
# model = wide_resnet.resnet50(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# k=opt.wide_resnet_k,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model == 'resnext':
# assert opt.model_depth in [50, 101, 152]
# from models.resnext import get_fine_tuning_parameters
# if opt.model_depth == 50:
# model = resnext.resnet50(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# cardinality=opt.resnext_cardinality,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 101:
# model = resnext.resnet101(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# cardinality=opt.resnext_cardinality,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 152:
# model = resnext.resnet152(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# cardinality=opt.resnext_cardinality,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model == 'preresnet':
# assert opt.model_depth in [18, 34, 50, 101, 152, 200]
# from models.pre_act_resnet import get_fine_tuning_parameters
# if opt.model_depth == 18:
# model = pre_act_resnet.resnet18(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 34:
# model = pre_act_resnet.resnet34(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 50:
# model = pre_act_resnet.resnet50(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 101:
# model = pre_act_resnet.resnet101(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 152:
# model = pre_act_resnet.resnet152(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 200:
# model = pre_act_resnet.resnet200(
# num_classes=opt.n_classes,
# shortcut_type=opt.resnet_shortcut,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model == 'densenet':
# assert opt.model_depth in [121, 169, 201, 264]
# from models.densenet import get_fine_tuning_parameters
# if opt.model_depth == 121:
# model = densenet.densenet121(
# num_classes=opt.n_classes,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 169:
# model = densenet.densenet169(
# num_classes=opt.n_classes,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 201:
# model = densenet.densenet201(
# num_classes=opt.n_classes,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
# elif opt.model_depth == 264:
# model = densenet.densenet264(
# num_classes=opt.n_classes,
# sample_size=opt.sample_size,
# sample_duration=opt.sample_duration)
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
if opt.model_depth == 50:
resnet = torchvision.models.resnet50(pretrained=True)
elif opt.model_depth ==101:
resnet = torchvision.models.resnet101(pretrained=True)
else:
raise Exception("Only resnet 50 and 101 pretrained on imagenet are provided now")
else:
raise Exception("currently only resnet50 and resnet101 are provided")
if not opt.no_cuda:
i3resnet = I3ResNet(copy.deepcopy(resnet), opt.sample_duration)
model = i3resnet.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
#assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in ['resnet', 'resnetl', 'resnext', 'c3d']
if opt.model == 'resnet':
assert opt.model_depth in [10]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnetl':
assert opt.model_depth in [10]
from models.resnetl import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnetl.resnetl10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [101]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'c3d':
assert opt.model_depth in [10]
from models.c3d import get_fine_tuning_parameters
if opt.model_depth == 10:
model = c3d.c3d_v1(sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
num_classes=opt.n_classes)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'c3d': # CHECK HERE
model.module.fc = nn.Linear(model.module.fc[0].in_features,
opt.n_finetune_classes)
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
if opt.modality == 'RGB' and opt.model != 'c3d':
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(
model, 3, 3) ##### Check models trained (3,7,7) or (7,7,7)
elif opt.modality == 'Depth':
print(
"[INFO]: Converting the pretrained model to Depth init model")
model = _construct_depth_model(model)
print("[INFO]: Done. Flow model ready.")
elif opt.modality == 'RGB-D':
print(
"[INFO]: Converting the pretrained model to RGB+D init model")
model = _construct_rgbdepth_model(model)
print("[INFO]: Done. RGB-D model ready.")
modules = list(model.modules())
first_conv_idx = list(
filter(lambda x: isinstance(modules[x], nn.Conv3d),
list(range(len(modules)))))[0]
conv_layer = modules[first_conv_idx]
if conv_layer.kernel_size[0] > opt.sample_duration:
model = _modify_first_conv_layer(model,
int(opt.sample_duration / 2), 1)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.modality == 'RGB' and opt.model != 'c3d':
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(model, 3, 3)
elif opt.modality == 'Depth':
print(
"[INFO]: Converting the pretrained model to Depth init model")
model = _construct_depth_model(model)
print("[INFO]: Deoth model ready.")
elif opt.modality == 'RGB-D':
print(
"[INFO]: Converting the pretrained model to RGB-D init model")
model = _construct_rgbdepth_model(model)
print("[INFO]: Done. RGB-D model ready.")
modules = list(model.modules())
first_conv_idx = list(
filter(lambda x: isinstance(modules[x], nn.Conv3d),
list(range(len(modules)))))[0]
conv_layer = modules[first_conv_idx]
if conv_layer.kernel_size[0] > opt.sample_duration:
print("[INFO]: RGB model is used for init model")
model = _modify_first_conv_layer(model,
int(opt.sample_duration / 2), 1)
if opt.model == 'c3d': # CHECK HERE
model.fc = nn.Linear(model.fc[0].in_features,
model.fc[0].out_features)
else:
model.fc = nn.Linear(model.fc.in_features, opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet','senet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'senet':
assert opt.model_depth in [50,101,152,154,5032,10132]
if opt.model_depth == 50:
model = senet.se_resnet50(num_classes = opt.n_classes, pretrained = None)
elif opt.model_depth == 101:
model = senet.se_resnet101(num_classes = opt.n_classes, pretrained = None)
elif opt.model_depth == 152:
model = senet.se_resnet152(num_classes = opt.n_classes, pretrained = None)
elif opt.model_depth == 154:
model = senet.senet154(num_classes = opt.n_classes, pretrained = None)
elif opt.model_depth == 5032:
model = senet.resnext50_32x4d(num_classes = opt.n_classes, pretrained = None)
elif opt.model_depth == 10132:
model = senet.se_resnext101_32x4d(num_classes = opt.n_classes, pretrained = None)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
#assert opt.arch == pretrain['arch']
#model.load_state_dict(pretrain['state_dict'])
pretrained_dict = pretrain['state_dict']
model_dict = model.state_dict()
pretrained_dict = {k: v for k, v in pretrained_dict.items() if k.find("module.fc") == -1}
model_dict.update(pretrained_dict)
model.load_state_dict(model_dict)
#model.load_state_dict(model_dict,strict=False)
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
elif opt.model == "senet":
model.module.last_linear = nn.Linear(model.module.last_linear.in_features, opt.n_finetune_classes)
model.module.last_linear = model.module.last_linear.cuda()
return model, model.parameters()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path, map_location='cpu')
assert opt.arch == pretrain['arch']
from collections import OrderedDict
new_state_dict = OrderedDict()
for k, v in pretrain['state_dict'].items():
name = k[7:] # remove `module.`
new_state_dict[name] = v
# load params
model.load_state_dict(new_state_dict)
#model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.mode in ['score', 'feature']
if opt.mode == 'score':
last_fc = True
elif opt.mode == 'feature':
last_fc = False
assert opt.model_name in ['resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet']
if opt.model_name == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
polices = resnet.get_fine_tuning_parameters(model, opt.ft_begin_index)
elif opt.model_name == 'wideresnet':
assert opt.model_depth in [50]
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut, k=opt.wide_resnet_k,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_name == 'resnext':
assert opt.model_depth in [50, 101, 152]
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut, cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut, cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut, cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_name == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(num_classes=opt.n_classes, shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_name == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size, sample_duration=opt.sample_duration,
last_fc=last_fc)
return model, polices
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet', 'i3d',
'i3dv2'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == "i3d":
from models.i3dpt import get_fine_tuning_parameters
model = i3dpt.I3D(num_classes=opt.n_classes, dropout_prob=0.5)
elif opt.model == "i3dv2":
from models.I3D_Pytorch import get_fine_tuning_parameters
model = I3D_Pytorch.I3D(num_classes=opt.n_classes,
dropout_keep_prob=0.5)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
if opt.model != "i3d" and opt.model != "i3dv2":
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
else:
pretrain = {"module." + k: v for k, v in pretrain.items()}
model_dict = model.state_dict()
model_dict.update(pretrain)
model.load_state_dict(model_dict)
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features,
opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(model.classifier.in_features,
opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
def generate_model(config):
assert config.base_model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
print(config.base_model, config.n_classes)
if config.base_model == 'resnet':
from models.resnet import get_fine_tuning_parameters
create_resnet_model = get_resnet_model(config.model_depth)
base_model = create_resnet_model(
num_classes=config.n_classes,
shortcut_type=config.resnet_shortcut,
sample_size=config.sample_size,
sample_duration=config.sample_duration)
elif config.base_model == 'wideresnet':
assert config.model_depth in [50], f'Wideresnet model depth {config.model_depth} is not valid.' \
f' Acceptable: {WIDERESNET_DEPTHS}'
from models.wide_resnet import get_fine_tuning_parameters
if config.model_depth == 50:
base_model = wide_resnet.resnet50(
num_classes=config.n_classes,
shortcut_type=config.resnet_shortcut,
k=config.wide_resnet_k,
sample_size=config.sample_size,
sample_duration=config.sample_duration)
elif config.base_model == 'resnext':
from models.resnext import get_fine_tuning_parameters
create_resnext_model = get_resnext_model(config.model_depth)
base_model = create_resnext_model(
num_classes=config.n_classes,
shortcut_type=config.resnet_shortcut,
cardinality=config.resnext_cardinality,
sample_size=config.sample_size,
sample_duration=config.sample_duration)
elif config.base_model == 'preresnet':
from models.pre_act_resnet import get_fine_tuning_parameters
create_preresnet_model = get_preresnet_model(config.model_depth)
base_model = create_preresnet_model(
num_classes=config.n_classes,
shortcut_type=config.resnet_shortcut,
sample_size=config.sample_size,
sample_duration=config.sample_duration)
elif config.base_model == 'densenet':
from models.densenet import get_fine_tuning_parameters
create_densenet_model = get_densenet_model(config.model_depth)
base_model = create_densenet_model(
num_classes=config.n_classes,
sample_size=config.sample_size,
sample_duration=config.sample_duration)
if config.cuda_available:
if config.cuda_id1 != -1:
base_model = nn.DataParallel(
base_model, device_ids=[config.cuda_id0, config.cuda_id1])
else:
base_model = nn.DataParallel(base_model,
device_ids=[config.cuda_id0])
if config.pretrain_path:
base_model = load_pretrained(base_model, config.base_model,
config.model_depth, config.pretrain_path)
if config.use_embeddings:
model = EmbeddingModel(base_model, config)
else:
model = update_with_finetune_block(base_model, config)
parameters = get_fine_tuning_parameters(
model, config.ft_begin_index
) # should not fail due to assert in the beginning
return model, parameters
def generate_model(opt):
assert opt.model in [
'resnet', 'resnext'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration,
isSource = opt.isSource,
transfer_module = opt.transfer_module,
sourceKind = opt.sourceKind,
layer_num = opt.layer_num,
multi_source = opt.multi_source)
print(opt.no_cuda)
print(type(opt.no_cuda))
if not opt.no_cuda:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
print('loading pretrained model arch', pretrain['arch'], opt.arch)
assert opt.arch == pretrain['arch']
pretrained_dict = pretrain['state_dict']
model_dict = model.state_dict()
# pretrained_dict = {k: v for k, v in pretrained_dict.items() if k in model_dict}
pretrained_dict = {str.replace(k,'module.',''): v for k,v in pretrained_dict.items()}
model_dict.update(pretrained_dict)
model.load_state_dict(model_dict)
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.inference == False:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
print(model)
return model, parameters
elif opt.inference:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
return model, model.parameters()
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
print('loading pretrained model arch', pretrain['arch'])
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
return model, parameters
return model, model.parameters()
def generate_model(opt):
assert opt.model in [
'resnet', 'preresnet', 'wideresnet', 'resnext', 'densenet'
]
if opt.model == 'resnet':
assert opt.model_depth in [10, 18, 34, 50, 101, 152, 200]
from models.resnet import get_fine_tuning_parameters
if opt.model_depth == 10:
model = resnet.resnet10(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 18:
model = resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'wideresnet':
assert opt.model_depth in [50]
from models.wide_resnet import get_fine_tuning_parameters
if opt.model_depth == 50:
model = wide_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
k=opt.wide_resnet_k,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'resnext':
assert opt.model_depth in [50, 101, 152]
from models.resnext import get_fine_tuning_parameters
if opt.model_depth == 50:
model = resnext.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = resnext.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = resnext.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
cardinality=opt.resnext_cardinality,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'preresnet':
assert opt.model_depth in [18, 34, 50, 101, 152, 200]
from models.pre_act_resnet import get_fine_tuning_parameters
if opt.model_depth == 18:
model = pre_act_resnet.resnet18(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 34:
model = pre_act_resnet.resnet34(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 50:
model = pre_act_resnet.resnet50(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 101:
model = pre_act_resnet.resnet101(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 152:
model = pre_act_resnet.resnet152(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 200:
model = pre_act_resnet.resnet200(
num_classes=opt.n_classes,
shortcut_type=opt.resnet_shortcut,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model == 'densenet':
assert opt.model_depth in [121, 169, 201, 264]
from models.densenet import get_fine_tuning_parameters
if opt.model_depth == 121:
model = densenet.densenet121(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 169:
model = densenet.densenet169(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 201:
model = densenet.densenet201(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
elif opt.model_depth == 264:
model = densenet.densenet264(
num_classes=opt.n_classes,
sample_size=opt.sample_size,
sample_duration=opt.sample_duration)
if not opt.no_cuda:
model = model.cuda()
model = nn.DataParallel(model, device_ids=None)
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.module.classifier = nn.Linear(
model.module.classifier.in_features, opt.n_finetune_classes)
model.module.classifier = model.module.classifier.cuda()
else:
model.module.fc = nn.Linear(model.module.fc.in_features,
opt.n_finetune_classes)
model.module.fc = model.module.fc.cuda()
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
else:
if opt.pretrain_path:
print('loading pretrained model {}'.format(opt.pretrain_path))
pretrain = torch.load(opt.pretrain_path)
assert opt.arch == pretrain['arch']
model.load_state_dict(pretrain['state_dict'])
if opt.model == 'densenet':
model.classifier = nn.Linear(
model.classifier.in_features, opt.n_finetune_classes)
else:
model.fc = nn.Linear(model.fc.in_features,
opt.n_finetune_classes)
parameters = get_fine_tuning_parameters(model, opt.ft_begin_index)
return model, parameters
return model, model.parameters()
