(opts, args) = parser.parse_args()
assert isinstance(opts, object)
opt = Config(opts.config)
mylog, logfile= get_logger(fileName=opt.log_name)
print(opt)
os.popen('cat {0} >> {1}'.format(opts.config, logfile))
if opt.checkpoint_folder is None:
opt.checkpoint_folder = 'models_checkpoint'
# make dir
if not os.path.exists(opt.checkpoint_folder):
os.system('mkdir {0}'.format(opt.checkpoint_folder))
train_dataset = dset(opt.data_dir, flist=opt.flist)
mylog.info('number of train samples is: {0}'.format(len(train_dataset)))
mylog.info('finished loading data')
os.environ['CUDA_VISIBLE_DEVICES'] = opt.gpu_id
ngpu = int(opt.ngpu)
opt.manualSeed = random.randint(1, 10000) # fix seed
# opt.manualSeed = 123456
if torch.cuda.is_available() and not opt.cuda:
mylog.info("WARNING: You have a CUDA device, so you should probably run with --cuda")
else:
if ngpu == 1:
mylog.info('so we use 1 gpu to training')
# forward, backward optimize
feat = deep_model.forward(vfeat_var, afeat_var)
X.append(feat.cpu().data.numpy())
cum_sample += 1
print('extract deep features: {} / {}'.format(cum_sample, num_sample))
X = np.concatenate(X)
y = np.concatenate(y)
print('extract deep features complete.')
return X, y
if __name__ == '__main__':
# load dataset
train_dataset = dset(bf_opt.data_dir, flist=bf_opt.flist)
print('number of train samples is: {0}'.format(len(train_dataset)))
print('finished loading data')
# train data loader
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=bf_opt.batchSize,
shuffle=True,
num_workers=int(bf_opt.workers))
# create neural network deep_model
deep_model = extract_models.VAMetric()
if bf_opt.init_deep_model != '':
print('loading pretrained deep_model from {0}'.format(
bf_opt.init_deep_model))
import setproctitle
setproctitle.setproctitle('train@changshuhao')
# reminding the cuda option
if torch.cuda.is_available():
if not opt.cuda:
print(
"WARNING: You have a CUDA device, so you should probably run with \"cuda: True\""
)
else:
cudnn.benchmark = True
# loading test dataset
test_video_dataset = dset(opt.data_dir, opt.video_flist, which_feat='vfeat')
test_audio_dataset = dset(opt.data_dir, opt.audio_flist, which_feat='afeat')
print('number of test samples is: {0}'.format(len(test_video_dataset)))
print('finished loading data')
# test function for metric learning
def test(video_loader, audio_loader, model, opt):
"""
train for one epoch on the training set
"""
# evaluation mode: only useful for the models with batchnorm or dropout
model.eval()
right = 0 # correct sample number
sample_num = 0 # total sample number
from torch.autograd import Variable
import numpy as np
import models_v3 as models
from dataset import VideoFeatDataset as dset
from tools import utils
# reminding the cuda option
if torch.cuda.is_available():
if not opt.cuda:
print("WARNING: You have a CUDA device, so you should probably run with \"cuda: True\"")
else:
cudnn.benchmark = True
# loading test dataset
test_video_dataset = dset(root=opt.data_dir, flist=opt.video_flist, which_feat='vfeat', creat_test=0)
test_audio_dataset = dset(root=opt.data_dir, flist=opt.audio_flist, which_feat='afeat', creat_test=0)
print('number of test samples is: {0}'.format(len(test_video_dataset)))
print('finished loading data')
# test function for metric learning
def test(video_loader, audio_loader, model, opt):
"""
train for one epoch on the training set
"""
# evaluation mode: only useful for the models with batchnorm or dropout
model.eval()
right = 0 # correct sample number
sample_num = 0 # total sample number
parser.add_argument("-tl", dest='transfer', help='transfer_learning', default=False, type=bool)
parser.add_argument("-c", dest='cfgfile', help="Config file",
default="cfg/yolov3.cfg", type=str)
parser.add_argument("-t", dest="use_tensorboard", help="Disable tensorboard", default=True, type=bool)
return parser.parse_args()
if __name__ == '__main__':
args = arg_parse()
# Use LMDB custom dataset or VOC-style
if cfg.lmdb:
dataset = dset(cfg.target_file, cfg.root_dir, cfg.multi_scale_inp_size) # , cfg.transforms)
dataloader = torch.utils.data.DataLoader(dataset, batch_size=args.batch, shuffle=True, num_workers=args.workers)
else:
image_data = torchvision.datasets.ImageFolder(args.path)
data_loader = torch.utils.data.DataLoader(image_data, batch_size=args.batch, shuffle=True, num_workers=args.workers, multiscale=cfg.multi_scale_inp_size)
# replace 4 with the number of classes in your custom dataset
classes = 20 if args.transfer else cfg.num_classes
# create the network
net = Darknet(classes)
# Load weights
