def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
roidb = get_training_roidb(imdb)
return roidb
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
roidb = get_training_roidb(imdb)
return roidb
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD) # 设置proposal方法，这里是selective search（config.py）
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
roidb = get_training_roidb(imdb) # 得到用于训练的roidb,定义在train.py,进行了水平翻转，以及为原始roidb添加了一些说明性的属性
return roidb
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)#得到对应数据集的对象,coco和pascal_voc都是imdb的子类
print('Loaded dataset `{:s}` for training'.format(imdb.name))#加载数据集 图片
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))#区域准则方法是gt,
roidb = get_training_roidb(imdb)#加载roi
return roidb
def combined_roidb(imdb_names):
"""
Combine multiple roidbs
"""
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
roidb = get_training_roidb(imdb)
return imdb, roidb
# roidbs = [get_roidb(s) for s in imdb_names.split('+')]
# imdb, roidb = roidbs[0]
imdb = get_imdb(imdb_names)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
roidb = get_training_roidb(imdb)
# if len(roidbs) > 1:
# for r in roidbs[1:]:
# roidb.extend(r)
# tmp = get_imdb(imdb_names.split('+')[1])
# imdb = datasets.imdb.imdb(imdb_names, tmp.classes)
# else:
# imdb = get_imdb(imdb_names)
return imdb, roidb
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name) # factory.py在工厂类中的get_imdb()
print('Loaded dataset `{:s}` for training'.format(imdb.name))
# 读取imdb后，执行set_proposal_method函数，该函数把字符串转成表达式
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
roidb = get_training_roidb(imdb)
return roidb
def combined_roidb(imdb_name):
"""
Combine multiple roidbs
"""
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
roidb = get_training_roidb(imdb)
#if cfg.TRAIN.USE_RPN_DB:
# roidb = imdb.add_rpn_rois()
return imdb, roidb
def get_roidb(imdb_name):
## imdb为pascal_voc类的一个对象
## 内容里有该类里的各种self名称与操作，包括roi信息等等
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
## __C.TRAIN.PROPOSAL_METHOD = 'gt'
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD)
print('Set proposal method: {:s}'.format(cfg.TRAIN.PROPOSAL_METHOD))
## get_training_roidb函数返回imdb对象的各种roi与图片信息，用于训练
## 这是一个列表，列表中存的是各个图片的字典，字典中存roi信息，字典引索为图片引索
roidb = get_training_roidb(imdb)
return roidb
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
imdb.set_proposal_method(cfg.TRAIN.PROPOSAL_METHOD) #这句是什么意思？？？？
print('Set proposal method: {:s}'.format(
cfg.TRAIN.PROPOSAL_METHOD)) ##cfg.TRAIN.PROPOSAL_METHOD='gt'
roidb = get_training_roidb(imdb)
'''
roidb是一个列表,长度为10022,(10022,),,每个元素是一个字典
imdb.roidb[0]:
{
'gt_classes': array([9, 9, 9], dtype=int32), 'max_classes': array([9, 9, 9]),
'image': '/home/beitadoge/.../000005.jpg', 'flipped': False, 'width': 500,
'boxes': array([[262, 210, 323, 338],[164, 263, 252, 371],[240, 193, 294, 298]], dtype=uint16),
'max_overlaps': array([ 1., 1., 1.], dtype=float32), 'height': 375,
'seg_areas': array([ 7998., 9701., 5830.], dtype=float32),
'gt_overlaps': <3x21 sparse matrix of type '<type 'numpy.float32'>'with 3 stored elements in Compressed Sparse Row format>
}
'''
return roidb
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('Using config:')
pprint.pprint(cfg)
np.random.seed(cfg.RNG_SEED)
# train set
# imdb, roidb = combined_roidb(args.imdb_name)
# print('{:d} roidb entries'.format(len(roidb)))
imdb = get_imdb(args.imdb_name)
roidb = get_training_roidb(imdb)
valroidb = roidb
# output directory where the models are saved
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 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
def get_roidb(imdb_name):
imdb = get_imdb(imdb_name)
print('Loaded dataset `{:s}` for training'.format(imdb.name))
roidb = get_training_roidb(imdb)
return roidb
print(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('Using config:')
pprint.pprint(cfg)
np.random.seed(cfg.RNG_SEED)
# train set
# imdb, roidb = combined_roidb(args.imdb_name)
imdb = get_imdb(args.imdb_name)
roidb = get_training_roidb(imdb)
print('{:d} roidb entries'.format(len(roidb)))
# output directory where the models are saved
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 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)))
