gpu_list = str(GPU)
else:
multi_gpus = True
for i, gpu_id in enumerate(GPU):
gpu_list += str(gpu_id)
if i != len(GPU) - 1:
gpu_list += ','
os.environ['CUDA_VISIBLE_DEVICES'] = gpu_list
# other init
start_epoch = 1
save_dir = os.path.join(SAVE_DIR, MODEL_PRE + 'v2_' + datetime.now().strftime('%Y%m%d_%H%M%S'))
if os.path.exists(save_dir):
raise NameError('model dir exists!')
os.makedirs(save_dir)
logging = init_log(save_dir)
_print = logging.info
# define trainloader and testloader
trainset = CASIA_Face(root=CASIA_DATA_DIR)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=BATCH_SIZE,
shuffle=True, num_workers=8, drop_last=False)
# nl: left_image_path
# nr: right_image_path
nl, nr, folds, flags = parseList(root=LFW_DATA_DIR)
testdataset = LFW(nl, nr)
testloader = torch.utils.data.DataLoader(testdataset, batch_size=32,
shuffle=False, num_workers=8, drop_last=False)
else:
multi_gpus = True
for i, gpu_id in enumerate(GPU):
gpu_list += str(gpu_id)
if i != len(GPU) - 1:
gpu_list += ','
os.environ['CUDA_VISIBLE_DEVICES'] = gpu_list
# other init
start_epoch = 1
save_dir = os.path.join(
SAVE_DIR, MODEL_PRE + 'v2_' + datetime.now().strftime('%Y%m%d_%H%M%S'))
if os.path.exists(save_dir):
raise NameError('model dir exists!')
os.makedirs(save_dir)
logging = init_log(save_dir)
_print = logging.info
# define trainloader and testloader
trainset = CASIA_Face(root=CASIA_DATA_DIR)
trainloader = torch.utils.data.DataLoader(trainset,
batch_size=BATCH_SIZE,
shuffle=True,
num_workers=8,
drop_last=False)
# nl: left_image_path
# nr: right_image_path
nl, nr, folds, flags = parseList(root=LFW_DATA_DIR)
testdataset = LFW(nl, nr)
testloader = torch.utils.data.DataLoader(testdataset,
parser.add_option("-e", "--trainingEpochs", action="store", type="int", dest="trainingEpochs", default=10, help="Number of training epochs")
parser.add_option("-b", "--batchSize", action="store", type="int", dest="batchSize", default=16, help="Batch Size")
#Input Reader Params
parser.add_option("--ft", action="store_true", dest="ft", default=False, help="Use pre-trained models from DermNet")
parser.add_option("--cutout", action="store_true", dest="cutout", default=False, help="applying cutout")
parser.add_option("--focal", action="store_true", dest="focal", default=False, help="applying focal loss")
# Parse command line options
(options, args) = parser.parse_args()
print(options)
start_epoch = 1
if os.path.exists(options.save_dir):
raise NameError('model dir exists!')
os.makedirs(options.save_dir)
logging = init_log(options.save_dir)
_print = logging.info
# read dataset
trainset = dataset.CUB(root_dir='core/2016train', is_train=True)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=options.batchSize,
shuffle=True, num_workers=0, drop_last=False)
testset = dataset.CUB(root_dir='core/2016test', is_train=False)
testloader = torch.utils.data.DataLoader(testset, batch_size=options.batchSize,
shuffle=False, num_workers=0, drop_last=False)
# define model
net = model.attention_net(topN=PROPOSAL_NUM)
if resume:
ckpt = torch.load(resume)
net.load_state_dict(ckpt['net_state_dict'])
start_epoch = ckpt['epoch'] + 1
