def get_netvlad():
checkpoint_path = Path(
EXPER_PATH,
'vd16_pitts30k_conv5_3_vlad_preL2_intra_white/vd16_pitts30k_conv5_3_vlad_preL2_intra_white'
)
config = {
'checkpoint_path':
checkpoint_path,
'data': {
'name': 'aachen',
'load_db': False,
'load_queries': True,
'resize_max': 960
},
'model': {
'name': 'netvlad_original',
'local_descriptor_layer': 'conv3_3',
'image_channels': 1
},
'weights':
'vd16_pitts30k_conv5_3_vlad_preL2_intra_white/vd16_pitts30k_conv5_3_vlad_preL2_intra_white'
}
net = get_model(config['model']['name'])(data_shape={
'image': [None, None, None, 3]
},
**config['model'])
net.load(str(checkpoint_path))
return net
def _init_graph(config, with_dataset=False):
set_seed(config.get('seed', int.from_bytes(os.urandom(4), byteorder='big')))
n_gpus = len(os.environ['CUDA_VISIBLE_DEVICES'].split(','))
logging.info('Number of GPUs detected: {}'.format(n_gpus))
dataset = get_dataset(config['data']['name'])(**config['data'])
model = get_model(config['model']['name'])(
data=dataset.get_tf_datasets(), n_gpus=n_gpus, **config['model'])
model.__enter__()
if with_dataset:
yield model, dataset
else:
yield model
model.__exit__()
tf.reset_default_graph()
with open(Path(EXPER_PATH, exper_name, 'config.yaml'), 'r') as f:
config['model'] = tools.dict_update(
yaml.load(f)['model'], config.get('model', {}))
checkpoint_path = Path(EXPER_PATH, exper_name)
if config.get('weights', None):
checkpoint_path = Path(checkpoint_path, config['weights'])
else:
if config.get('weights', None):
checkpoint_path = Path(DATA_PATH, 'weights', config['weights'])
else:
checkpoint_path = None
logging.info('No weights provided.')
logging.info(f'Starting export with configuration:\n{pformat(config)}')
with get_model(config['model']['name'])(data_shape={
'image': [None, None, None, config['model']['image_channels']]
},
**config['model']) as net:
if checkpoint_path is not None:
net.load(str(checkpoint_path))
dataset = get_dataset(config['data']['name'])(**config['data'])
test_set = dataset.get_test_set()
for data in tqdm(test_set):
predictions = net.predict(data, keys=keys)
predictions['input_shape'] = data['image'].shape
name = data['name'].decode('utf-8')
Path(base_dir,
Path(name).parent).mkdir(parents=True, exist_ok=True)
np.savez(Path(base_dir, '{}.npz'.format(name)), **predictions)
