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)
############################## begin #######################################
# train set
# imdb, roidb = combined_roidb(args.imdb_name)
imdb = get_Imdbs(args.imdb_name)
# total num_images
total_num = imdb.num_images
# initial num_images
initialnum = imdb[imdb.item_name(0)].num_images
# unlabeled num_images
remainnum = imdb[imdb.item_name(1)].num_images
print('total num:{}, initial num:{}'.format(total_num,initialnum))
bitmapImdb = BitMap(total_num)
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))
imdb = get_Imdbs(args.imdb_name)
roidb = get_training_roidb(imdb)
print '{:d} roidb entries'.format(len(roidb))
output_dir = get_output_dir(imdb)
print 'Output will be saved to `{:s}`'.format(output_dir)
# some statistic to record
alamount = 0
ssamount = 0
# set bitmap for AL
tableA = BitMap(imdb.num_images)
# choose initiail samples:VOC2007
sample_num = imdb.num_images
train_num = len(imdb[imdb.item_name(0)].roidb)
print 'All VOC2007 images use for initial train, image numbers:%d' % (
train_num)
for i in range(train_num):
tableA.set(i)
train_roidb = [roidb[i] for i in range(train_num)]
pretrained_model_name = args.pretrained_model
# static parameters
tao = 60000
beta = 1000
# updatable hypeparameters
gamma = 0.6
mylambda = np.array([-np.log(0.9)] * imdb.num_classes)
# train record
######################## begin #############################
imdb = get_Imdbs(args.imdb_name)
roidb = get_training_roidb(imdb)
print '{:d} roidb entries'.format(len(roidb))
output_dir = get_output_dir(imdb)
print 'Output will be saved to `{:s}`'.format(output_dir)
# some statistic to record
alamount = 0; ssamount = 0
discardamount = 0
# set bitmap for AL
bitmapImdb = BitMap(imdb.num_images)
# choose initiail samples:VOC2007
initial_num = len(imdb[imdb.item_name(0)].roidb)
print 'All VOC2007 images use for initial train, image numbers:%d'%(initial_num)
for i in range(initial_num):
bitmapImdb.set(i)
train_roidb = [roidb[i] for i in range(initial_num)]
pretrained_model_name = args.pretrained_model
# static parameters
tao = args.max_iters
# initial hypeparameters
gamma = 0.15; clslambda = np.array([-np.log(0.9)]*imdb.num_classes)
# train record
loopcounter = 0; train_iters = 0; iters_sum = train_iters
# control al proportion
al_proportion_checkpoint = [int(x*initial_num) for x in np.linspace(0.3,2,10)]