def set_config_params(args):
if args.cfg_file != '':
cfg_from_file(args.cfg_file)
if args.train is not None:
cfg.TRAIN.FLAG = args.train
if args.data_dir != '':
cfg.DATA_DIR = args.data_dir
cfg.DATA_DIR = os.path.join(cfg.DATA_DIR, cfg.DATA_SIZE)
# Set device
if torch.cuda.is_available():
cfg.DEVICE = torch.device('cuda')
if args.threshold is not None:
cfg.VOCAB.THRESHOLD = args.threshold
print('Using config:')
pprint.pprint(cfg)
cap_lens = cap_lens[sorted_indices]
cap_array = np.zeros((len(captions), max_len), dtype='int64')
for i in range(len(captions)):
idx = sorted_indices[i]
cap = captions[idx]
c_len = len(cap)
cap_array[i, :c_len] = cap
key = name[(name.rfind('/') + 1):]
data_dic[key] = [cap_array, cap_lens, sorted_indices]
algo.gen_example(data_dic)
if __name__ == "__main__":
args = parse_args()
if args.cfg_file is not None:
cfg_from_file(args.cfg_file)
if args.data_dir != '':
cfg.DATA_DIR = args.data_dir
print('Using config:')
pprint.pprint(cfg)
if not cfg.TRAIN.FLAG:
args.manualSeed = 100
elif args.manualSeed is None:
args.manualSeed = random.randint(1, 10000)
random.seed(args.manualSeed)
np.random.seed(args.manualSeed)
torch.manual_seed(args.manualSeed)
if cfg.CUDA:
torch.cuda.manual_seed_all(args.manualSeed)
cap = captions[idx]
c_len = len(cap)
cap_array[i, :c_len] = cap
key = name[(name.rfind('/') + 1):]
data_dic[key] = [cap_array, cap_lens, sorted_indices]
algo.gen_example(data_dic)
if __name__ == "__main__":
m = Main() # change. object of Main class to handle code execution
# args = parse_args() # argument parser object
# check if the passed config file is not None
if m.cfg_file is not None: # change
# cf_file = "/Users/nikunjlad/Github/MirrorGAN/cfg/train_bird.yml"
cfg_from_file(m.cfg_file)
# update default configs and user given configs
if m.data_dir != '':
cfg.DATA_DIR = m.data_dir
print('Using config:')
pprint.pprint(cfg)
# setting random seed fir numpy and torch tensors
if not cfg.TRAIN.FLAG:
m.manualSeed = 100
elif m.manualSeed is None:
m.manualSeed = random.randint(1, 10000)
random.seed(m.manualSeed)
np.random.seed(m.manualSeed)
torch.manual_seed(m.manualSeed)
