def __init__(self, i_net):
# Create network
self.netname = i_net
print('Train: cfg.FLAGS.network', net)
if i_net == 'vgg16':
self.net = vgg16(batch_size=cfg.FLAGS.ims_per_batch)
self.pre_model = cfg.FLAGS.pretrained_model
elif i_net == 'res50':
self.net = resnetv1(num_layers=50)
self.pre_model = cfg.FLAGS.pretrained_resnet50_model
elif i_net == 'res101':
self.net = resnetv1(num_layers=101)
self.pre_model = cfg.FLAGS.pretrained_resnet101_model
elif i_net == 'res152':
self.net = resnetv1(num_layers=152)
self.pre_model = cfg.FLAGS.pretrained_resnet152_model
else:
raise NotImplementedError
print('self.net', self.net)
self.imdb, self.roidb = combined_roidb("voc_2007_trainval")
self.data_layer = RoIDataLayer(self.roidb, self.imdb.num_classes)
self.output_dir = cfg.get_output_dir(self.imdb, 'default')
def setup():
global net, vocab
ckpt_dir = 'output/ckpt/'
vocabulary = 'output/ckpt/vocabulary.txt'
# load network
net = resnetv1(num_layers=50) # vgg16() resnetv1(num_layers=50, 101, 152)
net.create_architecture("TEST", num_classes=1, tag='pre')
ckpt = tf.train.get_checkpoint_state(ckpt_dir)
# load vocab
vocab = ['<PAD>', '<SOS>', '<EOS>']
with open(vocabulary, 'r') as f:
for line in f:
vocab.append(line.strip())
# set config
tfconfig = tf.ConfigProto(allow_soft_placement=True)
tfconfig.gpu_options.allow_growth = True
# init session
saver = tf.train.Saver()
sess = tf.Session(config=tfconfig)
saver.restore(sess, ckpt.model_checkpoint_path)
print('Restored from {}'.format(ckpt.model_checkpoint_path))
return sess
def architecture_test():
blob = get_data_test()
tf.reset_default_graph()
net = resnetv1(50)
# net._build_network()
net.create_architecture(mode='TEST', tag='pre')
for n in tf.get_default_graph().as_graph_def().node:
print(n.name)
tfconfig = tf.ConfigProto(allow_soft_placement=True)
tfconfig.gpu_options.allow_growth = True
feed_dict = {
net._image: blob['data'],
net._im_info: blob['im_info'],
net._gt_boxes: blob['gt_boxes'],
net._gt_phrases: blob['gt_phrases']
}
output = net._for_debug
output.update({
"image": net._image,
"im_info": net._im_info,
"gt_boxes": net._gt_boxes,
"gt_phrases": net._gt_phrases
})
with tf.Session(config=tfconfig) as sess:
init = tf.global_variables_initializer()
sess.run(init)
out = sess.run('DenseCap_ResNet50/Prediction/lstm/cap_init_state:0',
feed_dict=feed_dict)
print(out.shape)
def __init__(self):
# Create network
if cfg.FLAGS.network == 'resnet_v1':
self.net = resnetv1(batch_size=cfg.FLAGS.ims_per_batch,
num_layers=50)
else:
raise NotImplementedError
self.imdb, self.roidb = combined_roidb("voc_2007_trainval")
self.data_layer = RoIDataLayer(self.roidb, self.imdb.num_classes)
self.output_dir = cfg.get_output_dir(self.imdb, 'default')
def __init__(self):
# Create network
if cfg.FLAGS.net == 'vgg16':
self.net = vgg16(batch_size=cfg.FLAGS.ims_per_batch)
elif cfg.FLAGS.net == 'resnetv1':
self.net = resnetv1(batch_size=cfg.FLAGS.ims_per_batch)
else:
raise NotImplementedError
self.imdb, self.roidb = combined_roidb("Columbia")
self.data_layer = RoIDataLayer(self.roidb, self.imdb.num_classes)
self.output_dir = cfg.get_output_dir(self.imdb, 'default')
def __init__(self, dataset):
# Create network
if cfg.FLAGS.network == 'vgg16':
self.net = vgg16(batch_size=cfg.FLAGS.ims_per_batch)
elif cfg.FLAGS.network == 'RESNET_v1_50':
self.net = resnetv1(batch_size=cfg.FLAGS.ims_per_batch)
else:
raise NotImplementedError
#The database
#self.imdb, self.roidb = combined_roidb("voc_2007_trainval+test+Isabel")
self.imdb, self.roidb = combined_roidb(dataset)
#self.imdb, self.roidb = combined_roidb("Isabel")
print(self.imdb.name)
self.data_layer = RoIDataLayer(self.roidb, self.imdb.num_classes)
self.output_dir = cfg.get_output_dir(self.imdb, 'default')
print(self.output_dir)
print(tfmodel)
raise IOError(('{:s} not found.\nDid you download the proper networks from '
'our server and place them properly?').format(tfmodel + '.meta')
"""
# set config
tfconfig = tf.ConfigProto(allow_soft_placement=True)
tfconfig.gpu_options.allow_growth = True
# init session
sess = tf.Session(config=tfconfig)
# load network
if demonet == 'vgg16':
net = vgg16(batch_size=1)
elif demonet == 'res101':
net = resnetv1(batch_size=1, num_layers=101)
else:
raise NotImplementedError
net.create_architecture(sess,
"TEST",
21,
tag='default',
anchor_scales=[8, 16, 32])
saver = tf.train.Saver()
# read from the pre-store ckpt file
saver.restore(sess, tfmodel)
print('Loaded network {:s}'.format(tfmodel))
im_names = [
# tensorboard directory where the summaries are saved during training
tb_dir = get_output_tb_dir(imdb, args.tag)
print('TensorFlow summaries will be saved to `{:s}`'.format(tb_dir))
# also add the validation set, but with no flipping images
orgflip = cfg.TRAIN.USE_FLIPPED
cfg.TRAIN.USE_FLIPPED = False
_, valroidb = combined_roidb(args.imdbval_name)
print('{:d} validation roidb entries'.format(len(valroidb)))
cfg.TRAIN.USE_FLIPPED = orgflip
# load network
if args.net == 'vgg16':
net = vgg16(batch_size=cfg.TRAIN.IMS_PER_BATCH)
elif args.net == 'res50':
net = resnetv1(batch_size=cfg.TRAIN.IMS_PER_BATCH, num_layers=50)
elif args.net == 'res101':
net = resnetv1(batch_size=cfg.TRAIN.IMS_PER_BATCH, num_layers=101)
elif args.net == 'res152':
net = resnetv1(batch_size=cfg.TRAIN.IMS_PER_BATCH, num_layers=152)
else:
raise NotImplementedError
train_net(net,
imdb,
roidb,
valroidb,
output_dir,
tb_dir,
pretrained_model=args.weight,
max_iters=args.max_iters)
if not os.path.isfile(tfmodel + '.meta'):
raise IOError(
('{:s} not found.\nDid you download the proper networks from '
'our server and place them properly?').format(tfmodel + '.meta'))
# set config
tfconfig = tf.ConfigProto(allow_soft_placement=True)
tfconfig.gpu_options.allow_growth = True
# init session
sess = tf.Session(config=tfconfig)
# load network
if demonet == 'vgg16':
net = vgg16()
elif demonet == 'res101':
net = resnetv1(num_layers=101)
else:
raise NotImplementedError
net.create_architecture("TEST",
21,
tag='default',
anchor_scales=[8, 16, 32])
saver = tf.train.Saver()
saver.restore(sess, tfmodel)
print('Loaded network {:s}'.format(tfmodel))
im_names = [
'000456.jpg', '000542.jpg', '001150.jpg', '001763.jpg', '004545.jpg',
'dog_cat00.jpg', 'dog_cat02.jpg', 'dog_cat03.jpg'
]
if __name__ == '__main__':
args = parse_args()
print('------- called with args: --------')
pprint.pprint(args)
if args.cfg_file is not None:
cfg_from_file(args.cfg_file)
if args.set_cfgs is not None:
cfg_from_list(args.set_cfgs)
# load network
if args.net == 'vgg16':
net = vgg16()
elif args.net == 'res50':
net = resnetv1(num_layers=50)
elif args.net == 'res101':
net = resnetv1(num_layers=101)
elif args.net == 'res152':
net = resnetv1(num_layers=152)
else:
raise NotImplementedError
net.create_architecture("TEST", num_classes=1, tag='pre')
vocab = ['<PAD>', '<SOS>', '<EOS>']
with open(args.vocabulary, 'r') as f:
for line in f:
vocab.append(line.strip())
# get the image paths
im_paths = glob.glob('./data/demo/*.jpg')
def get_model():
args = parse_args()
# model path
demonet = args.demo_net
dataset = args.dataset
tfmodel = os.path.join('default', 'Columbia', 'default', NETS[demonet][0])
if not os.path.isfile(tfmodel + '.meta'):
print(tfmodel)
raise IOError(
('{:s} not found.\nDid you download the proper networks from '
'our server and place them properly?').format(tfmodel + '.meta'))
# set config
tfconfig = tf.ConfigProto(allow_soft_placement=True)
tfconfig.gpu_options.allow_growth = True
# init session
sess = tf.Session(config=tfconfig)
# load network
if demonet == 'vgg16':
net = vgg16(batch_size=1)
elif demonet == 'res50':
net = resnetv1(batch_size=1, num_layers=50)
else:
raise NotImplementedError
#net.create_architecture(sess, "TEST", 2, tag='default', anchor_scales=[8, 16, 32])
srm2 = tf.get_variable('vgg_16/srm2/weights', [5, 5, 3, 3],
trainable=False)
srm3 = tf.get_variable('vgg_16/srm3/weights', [5, 5, 3, 3],
trainable=False)
#srm2 = tf.get_variable('resnet_v1_50/srm/weights', [5, 5, 3, 3], trainable=False)
saver = tf.train.Saver()
saver.restore(sess, tfmodel)
#variables = tf.global_variables()
#for i in variables:
# print(i)
filters = sess.run(srm2)
print(filters.shape)
plt.imsave(
'filt0.jpg',
cv2.resize(filters[:, :, :, 0], (800, 800),
interpolation=cv2.INTER_NEAREST))
plt.imsave(
'filt1.jpg',
cv2.resize(filters[:, :, :, 1], (800, 800),
interpolation=cv2.INTER_NEAREST))
plt.imsave(
'filt2.jpg',
cv2.resize(filters[:, :, :, 2], (800, 800),
interpolation=cv2.INTER_NEAREST))
filters2 = sess.run(srm3)
print(filters2.shape)
plt.imsave(
'filt3.jpg',
cv2.resize(filters2[:, :, :, 0], (800, 800),
interpolation=cv2.INTER_NEAREST))
plt.imsave(
'filt4.jpg',
cv2.resize(filters2[:, :, :, 1], (800, 800),
interpolation=cv2.INTER_NEAREST))
plt.imsave(
'filt5.jpg',
cv2.resize(filters2[:, :, :, 2], (800, 800),
interpolation=cv2.INTER_NEAREST))
def main():
args = parse_args()
cfg.DATA_DIR = args.data_dir
cfg.CONTEXT_FUSION = args.context_fusion
# c_time = time.strftime('%m%d_%H%M', time.localtime())
# if not os.path.exists(cfg.LOG_DIR):
# os.makedirs(cfg.LOG_DIR)
# file_handler = logging.FileHandler(pjoin(cfg.LOG_DIR,
# args.network_name + '_%s.txt' % c_time))
# logging.getLogger().addHandler(file_handler)
print('------ called with args: -------')
pprint.pprint(args)
if args.cfg_file is not None:
cfg_from_file(args.cfg_file)
if args.set_cfgs is not None:
cfg_from_list(args.set_cfgs)
if cfg.INIT_BY_GLOVE and cfg.KEEP_AS_GLOVE_DIM:
cfg.EMBED_DIM = cfg.GLOVE_DIM
else:
cfg.EMBED_DIM = args.embed_dim
print("Using config:")
pprint.pprint(cfg)
if not args.randomize:
# fix the random seeds (numpy and caffe) for reproducibility
np.random.seed(cfg.RNG_SEED)
tf.set_random_seed(cfg.RNG_SEED)
imdb, roidb = GetRoidb(args.imdb_name)
output_dir = get_output_dir(imdb, args.tag)
print("output will be saved to `{:s}`".format(output_dir))
# tensorboard directory where the summaries are saved during training
tb_dir = get_output_tb_dir(imdb, args.tag)
print('TensorFlow summaries will be saved to `{:s}`'.format(tb_dir))
# also add validation set, but with no flipping image
orgflip = cfg.TRAIN.USE_FLIPPED
cfg.TRAIN.USE_FLIPPED = False
_, valroidb = GetRoidb(args.imdbval_name)
cfg.TRAIN.USE_FLIPPED = orgflip
# load network
if args.net == 'vgg16':
net = vgg16()
elif args.net == 'res50':
net = resnetv1(num_layers=50)
elif args.net == 'res101':
net = resnetv1(num_layers=101)
elif args.net == 'res152':
net = resnetv1(num_layers=152)
else:
raise NotImplementedError
if args.weights and not args.weights.endswith('.ckpt'):
try:
ckpt = tf.train.get_checkpoint_state(args.weights)
pretrained_model = ckpt.model_checkpoint_path
except:
raise ValueError("NO checkpoint found in {}".format(args.weights))
else:
pretrained_model = args.weights
# TODO: "imdb" may not be useful during training
train_net(net, imdb, roidb, valroidb, output_dir, tb_dir,
pretrained_model=pretrained_model,
max_iters=args.max_iters)
def main():
args = parse_args()
# c_time = time.strftime('%m%d_%H%M', time.localtime())
# if not os.path.exists(cfg.LOG_DIR):
# os.makedirs(cfg.LOG_DIR)
# file_handler = logging.FileHandler(pjoin(cfg.LOG_DIR,
# args.network_name + '_%s.txt' % c_time))
# logging.getLogger().addHandler(file_handler)
print('------ called with args: -------')
pprint.pprint(args)
if args.cfg_file is not None:
cfg_from_file(args.cfg_file)
if args.set_cfgs is not None:
cfg_from_list(args.set_cfgs)
print("runing with LIMIT_RAM: {}".format(cfg.LIMIT_RAM))
print("Using config:")
pprint.pprint(cfg)
if not args.randomize:
# fix the random seeds (numpy and caffe) for reproducibility
np.random.seed(cfg.RNG_SEED)
tf.set_random_seed(cfg.RNG_SEED)
if not cfg.LIMIT_RAM:
imdb, roidb = combined_roidb(args.imdb_name)
else:
imdb, roidb = get_roidb_limit_ram(args.imdb_name)
output_dir = get_output_dir(imdb, args.tag)
print("output will be saved to `{:s}`".format(output_dir))
# tensorboard directory where the summaries are saved during training
tb_dir = get_output_tb_dir(imdb, args.tag)
print('TensorFlow summaries will be saved to `{:s}`'.format(tb_dir))
# also add validation set, but with no flipping image
orgflip = cfg.TRAIN.USE_FLIPPED
cfg.TRAIN.USE_FLIPPED = False
if not cfg.LIMIT_RAM:
_, valroidb = combined_roidb(args.imdbval_name)
else:
_, valroidb = get_roidb_limit_ram(args.imdbval_name)
cfg.TRAIN.USE_FLIPPED = orgflip
# load network
if args.net == 'vgg16':
net = vgg16()
elif args.net == 'res50':
net = resnetv1(num_layers=50)
elif args.net == 'res101':
net = resnetv1(num_layers=101)
elif args.net == 'res152':
net = resnetv1(num_layers=152)
else:
raise NotImplementedError
# TODO: "imdb" may not be useful during training
train_net(net, imdb, roidb, valroidb, output_dir, tb_dir,
pretrained_model=args.pretrained_model,
max_iters=args.max_iters)
