def eval(num_imgs, eval_dir, annotation_dir, showbox):
test_imgname_list = [
item for item in os.listdir(eval_dir)
if item.endswith(('.jpg', 'jpeg', '.png', '.tif', '.tiff'))
]
if num_imgs == np.inf:
real_test_imgname_list = test_imgname_list
else:
real_test_imgname_list = test_imgname_list[:num_imgs]
fcos = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
all_boxes = eval_with_plac(det_net=fcos,
real_test_imgname_list=real_test_imgname_list,
img_root=eval_dir,
draw_imgs=showbox)
# save_dir = os.path.join(cfgs.EVALUATE_DIR, cfgs.VERSION)
# if not os.path.exists(save_dir):
# os.makedirs(save_dir)
# with open(os.path.join(save_dir, 'detections.pkl')) as f:
# all_boxes = pickle.load(f)
#
# print(len(all_boxes))
voc_eval.voc_evaluate_detections(all_boxes=all_boxes,
test_annotation_path=annotation_dir,
test_imgid_list=real_test_imgname_list)
def eval(num_imgs, eval_dir, annotation_dir, showbox):
# with open('/home/yjr/DataSet/VOC/VOC_test/VOC2007/ImageSets/Main/aeroplane_test.txt') as f:
# all_lines = f.readlines()
# test_imgname_list = [a_line.split()[0].strip() for a_line in all_lines]
test_imgname_list = [
item for item in os.listdir(eval_dir)
if item.endswith(('.jpg', 'jpeg', '.png', '.tif', '.tiff'))
]
if num_imgs == np.inf:
real_test_imgname_list = test_imgname_list
else:
real_test_imgname_list = test_imgname_list[:num_imgs]
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
eval_with_plac(det_net=faster_rcnn,
real_test_imgname_list=real_test_imgname_list,
img_root=eval_dir,
draw_imgs=showbox)
save_dir = os.path.join(cfgs.EVALUATE_DIR, cfgs.VERSION)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
with open(os.path.join(save_dir, 'detections.pkl'), 'rb') as f:
all_boxes = pickle.load(f)
print(len(all_boxes))
voc_eval.voc_evaluate_detections(all_boxes=all_boxes,
test_annotation_path=annotation_dir,
test_imgid_list=real_test_imgname_list)
def test(dataset, inference_save_path):
test_image_index = dataset.image_index
test_image_list=[dataset.image_path_from_index(index) for index in test_image_index]
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
detect(det_net=faster_rcnn, inference_save_path=inference_save_path, test_imgname_list=test_image_list)
def eval(img_dir, image_ext, test_annotation_path):
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
eval_with_plac(img_dir=img_dir,
det_net=faster_rcnn,
image_ext=image_ext,
draw_imgs=True)
with open(cfgs.VERSION + '_detections_h.pkl') as f1:
all_boxes_h = pickle.load(f1)
print(len(all_boxes_h))
with open(cfgs.VERSION + '_detections_r.pkl') as f2:
all_boxes_r = pickle.load(f2)
print(len(all_boxes_r))
imgs = os.listdir(img_dir)
real_test_imgname_list = [i.split(image_ext)[0] for i in imgs]
print(10 * "**")
print('horizon eval:')
# voc_eval.voc_evaluate_detections(all_boxes=all_boxes_h,
# test_imgid_list=real_test_imgname_list,
# test_annotation_path=test_annotation_path)
print(10 * "**")
print('rotation eval:')
voc_eval_r.voc_evaluate_detections(
all_boxes=all_boxes_r,
test_imgid_list=real_test_imgname_list,
test_annotation_path=test_annotation_path)
def eval(num_imgs, args):
if not args.show_box:
if not os.path.exists('./tmp.txt'):
fw = open('./tmp.txt', 'w')
fw.close()
fr = open('./tmp.txt', 'r')
img_filter = fr.readlines()
print('****************************'*3)
print('Already tested imgs:', img_filter)
print('****************************'*3)
fr.close()
test_imgname_list = [os.path.join(args.test_dir, img_name) for img_name in os.listdir(args.test_dir)
if img_name.endswith(('.jpg', '.png', '.jpeg', '.tif', '.tiff')) and
(img_name + '\n' not in img_filter)]
else:
test_imgname_list = [os.path.join(args.test_dir, img_name) for img_name in os.listdir(args.test_dir)
if img_name.endswith(('.jpg', '.png', '.jpeg', '.tif', '.tiff'))]
assert len(test_imgname_list) != 0, 'test_dir has no imgs there.' \
' Note that, we only support img format of (.jpg, .png, and .tiff) '
if num_imgs == np.inf:
real_test_img_list = test_imgname_list
else:
real_test_img_list = test_imgname_list[: num_imgs]
retinanet = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
test_dota(det_net=retinanet, real_test_img_list=real_test_img_list, args=args)
if not args.show_box:
os.remove('./tmp.txt')
def eval(num_imgs):
# annotation_path = '/home/yjr/DataSet/COCO/2017/test_annotations/image_info_test2017.json'
annotation_path = '/home/yjr/DataSet/COCO/2017/test_annotations/image_info_test-dev2017.json'
# annotation_path = '/home/yjr/DataSet/COCO/2017/annotations/instances_train2017.json'
print("load coco .... it will cost about 17s..")
coco = COCO(annotation_path)
imgId_list = coco.getImgIds()
if num_imgs != np.inf:
imgId_list = imgId_list[:num_imgs]
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
save_dir = os.path.join(cfgs.EVALUATE_DIR, cfgs.VERSION)
eval_with_plac(det_net=faster_rcnn,
coco=coco,
imgId_list=imgId_list,
out_json_root=save_dir,
draw_imgs=True)
print("each img over**************")
final_detections = []
with open(os.path.join(save_dir, 'coco2017test_results.json'), 'w') as wf:
for imgid in imgId_list:
f = open(os.path.join(save_dir, 'each_img', str(imgid) + '.json'))
tmp_list = json.load(f)
# print (type(tmp_list))
final_detections.extend(tmp_list)
del tmp_list
f.close()
json.dump(final_detections, wf)
def eval(num_imgs, img_dir, image_ext, test_annotation_path, draw_imgs):
retinanet = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
all_boxes_r = eval_with_plac(img_dir=img_dir,
det_net=retinanet,
num_imgs=num_imgs,
image_ext=image_ext,
draw_imgs=draw_imgs)
# with open(cfgs.VERSION + '_detections_r.pkl', 'rb') as f2:
# all_boxes_r = pickle.load(f2)
#
# print(len(all_boxes_r))
imgs = os.listdir(img_dir)
real_test_imgname_list = [i.split(image_ext)[0] for i in imgs]
print(10 * "**")
print('rotation eval:')
voc_eval_r.voc_evaluate_detections(
all_boxes=all_boxes_r,
test_imgid_list=real_test_imgname_list,
test_annotation_path=test_annotation_path)
def inference(src_dir, res_dir, draw_imgs):
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
detect(det_net=faster_rcnn,
src_dir=src_dir,
res_dir=res_dir,
draw_imgs=draw_imgs)
def inference(test_dir, inference_save_path):
test_imgname_list = [os.path.join(test_dir, img_name) for img_name in os.listdir(test_dir)
if img_name.endswith(('.jpg', '.png', '.jpeg', '.tif', '.tiff'))]
assert len(test_imgname_list) != 0, 'test_dir has no imgs there.' \
' Note that, we only support img format of (.jpg, .png, and .tiff) '
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
detect(det_net=faster_rcnn, inference_save_path=inference_save_path, real_test_imgname_list=test_imgname_list)
def eval(num_imgs, img_dir, image_ext, test_annotation_path):
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
eval_with_plac(img_dir=img_dir,
det_net=faster_rcnn,
num_imgs=num_imgs,
image_ext=image_ext,
draw_imgs=True,
test_annotation_path=test_annotation_path)
def eval(num_imgs, eval_data, json_file, showbox):
with open(json_file) as f:
test_img_list = json.load(f)['images']
if num_imgs == np.inf:
real_test_img_list = test_img_list
else:
real_test_img_list = test_img_list[: num_imgs]
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
test_coco(det_net=faster_rcnn, real_test_img_list=real_test_img_list, eval_data=eval_data, draw_imgs=showbox)
def eval(num_imgs, eval_dir, showbox):
test_imgname_list = [item for item in os.listdir(eval_dir)
if item.endswith(('.jpg', 'jpeg', '.png', '.tif', '.tiff'))]
if num_imgs == np.inf:
real_test_imgname_list = test_imgname_list
else:
real_test_imgname_list = test_imgname_list[: num_imgs]
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
eval_with_plac(det_net=faster_rcnn, real_test_imgname_list=real_test_imgname_list,
img_root=eval_dir,
draw_imgs=showbox)
def build_detection_graph():
# 1. preprocess img
img_plac = tf.placeholder(dtype=tf.uint8,
shape=[None, None, 3],
name='input_img') # is RGB. not GBR
raw_shape = tf.shape(img_plac)
raw_h, raw_w = tf.to_float(raw_shape[0]), tf.to_float(raw_shape[1])
img_batch = tf.cast(img_plac, tf.float32)
img_batch = short_side_resize_for_inference_data(
img_tensor=img_batch,
target_shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
length_limitation=cfgs.IMG_MAX_LENGTH)
if cfgs.NET_NAME in ['resnet101_v1d', 'resnet50_v1d']:
img_batch = (img_batch / 255 - tf.constant(
cfgs.PIXEL_MEAN_)) / tf.constant(cfgs.PIXEL_STD)
else:
img_batch = img_batch - tf.constant(cfgs.PIXEL_MEAN)
img_batch = tf.expand_dims(img_batch, axis=0) # [1, None, None, 3]
det_net = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
detected_boxes, detection_scores, detection_category = det_net.build_whole_detection_network(
input_img_batch=img_batch, gtboxes_batch=None)
xmin, ymin, xmax, ymax = detected_boxes[:, 0], detected_boxes[:, 1], \
detected_boxes[:, 2], detected_boxes[:, 3]
resized_shape = tf.shape(img_batch)
resized_h, resized_w = tf.to_float(resized_shape[1]), tf.to_float(
resized_shape[2])
xmin = xmin * raw_w / resized_w
xmax = xmax * raw_w / resized_w
ymin = ymin * raw_h / resized_h
ymax = ymax * raw_h / resized_h
boxes = tf.transpose(tf.stack([xmin, ymin, xmax, ymax]))
dets = tf.concat([
tf.reshape(detection_category, [-1, 1]),
tf.reshape(detection_scores, [-1, 1]), boxes
],
axis=1,
name='DetResults')
return dets
def eval(num_imgs, test_dir, gpu_ids, show_box):
test_imgname_list = [os.path.join(test_dir, img_name) for img_name in os.listdir(test_dir)
if img_name.endswith(('.jpg', '.png', '.jpeg', '.tif', '.tiff'))]
assert len(test_imgname_list) != 0, 'test_dir has no imgs there.' \
' Note that, we only support img format of (.jpg, .png, and .tiff) '
if num_imgs == np.inf:
real_test_img_list = test_imgname_list
else:
real_test_img_list = test_imgname_list[: num_imgs]
retinanet = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
test_icdar2015(det_net=retinanet, real_test_img_list=real_test_img_list, gpu_ids=gpu_ids, show_box=show_box)
def eval(num_imgs, eval_data, eval_gt, gpu_ids):
with open(eval_data) as f:
test_img_list = f.readlines()
if num_imgs == np.inf:
real_test_img_list = test_img_list
else:
real_test_img_list = test_img_list[: num_imgs]
fcos = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=False)
detected_json = eval_coco(det_net=fcos, real_test_img_list=real_test_img_list, gpu_ids=gpu_ids)
# save_path = os.path.join('./eval_coco', cfgs.VERSION)
# detected_json = os.path.join(save_path, 'coco_minival.json')
cocoval(detected_json, eval_gt)
def eval(num_imgs, eval_data, eval_gt, showbox):
with open(eval_data) as f:
test_img_list = f.readlines()
if num_imgs == np.inf:
real_test_img_list = test_img_list
else:
real_test_img_list = test_img_list[:num_imgs]
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False, batch_size=1)
detected_json = eval_coco(det_net=faster_rcnn,
real_test_img_list=real_test_img_list,
draw_imgs=showbox)
# save_path = os.path.join('./eval_coco', cfgs.VERSION)
# detected_json = os.path.join(save_path, 'coco_minival.json')
cocoval(detected_json, eval_gt)
def __init__(self, gpu=0):
self.init_op = None
self.restorer = None
self.restore_ckpt = None
self.config = None
self.sess = None
self.gpu = str(gpu)
self.det_net = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
self.img_plac = None
self.img_batch = None
self.det_boxes_h = None
self.det_scores_h = None
self.det_category_h = None
self.det_boxes_r = None
self.det_scores_r = None
self.det_category_r = None
self.init_model()
def train():
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
# list as many types of layers as possible, even if they are not used now
with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane, \
slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
final_bbox, final_scores, final_category, loss_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch, gtboxes_batch=gtboxes_and_label)
# ----------------------------------------------------------------------------------------------------build loss
weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
rpn_location_loss = loss_dict['rpn_loc_loss']
rpn_cls_loss = loss_dict['rpn_cls_loss']
rpn_total_loss = rpn_location_loss + rpn_cls_loss
fastrcnn_cls_loss = loss_dict['fastrcnn_cls_loss']
fastrcnn_loc_loss = loss_dict['fastrcnn_loc_loss']
fastrcnn_total_loss = fastrcnn_cls_loss + fastrcnn_loc_loss
total_loss = rpn_total_loss + fastrcnn_total_loss + weight_decay_loss
# ____________________________________________________________________________________________________build loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RPN_LOSS/cls_loss', rpn_cls_loss)
tf.summary.scalar('RPN_LOSS/location_loss', rpn_location_loss)
tf.summary.scalar('RPN_LOSS/rpn_total_loss', rpn_total_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss', fastrcnn_cls_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss', fastrcnn_loc_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_total_loss', fastrcnn_total_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(
img_batch=img_batch,
boxes=gtboxes_and_label[:, :-1],
labels=gtboxes_and_label[:, -1])
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = show_box_in_tensor.draw_boxes_with_categories_and_scores(
img_batch=img_batch,
boxes=final_bbox,
labels=final_category,
scores=final_scores)
tf.summary.image('Compare/final_detection', detections_in_img)
tf.summary.image('Compare/gtboxes', gtboxes_in_img)
# ___________________________________________________________________________________________________add summary
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(
global_step,
boundaries=[
np.int64(cfgs.DECAY_STEP[0]),
np.int64(cfgs.DECAY_STEP[1])
],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
# optimizer = tf.train.AdamOptimizer(lr)
# ---------------------------------------------------------------------------------------------compute gradients
gradients = faster_rcnn.get_gradients(optimizer, total_loss)
# enlarge_gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = faster_rcnn.enlarge_gradients_for_bias(gradients)
if cfgs.GRADIENT_CLIPPING_BY_NORM:
with tf.name_scope('clip_gradients_YJR'):
gradients = slim.learning.clip_gradient_norms(
gradients, cfgs.GRADIENT_CLIPPING_BY_NORM)
# _____________________________________________________________________________________________compute gradients
# train_op
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
summary_op = tf.summary.merge_all()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=30)
# Create session
#allow_soft_placement=True自动将无法放到GPU上的操作放回到CPU
config = tf.ConfigProto(allow_soft_placement=True)
#让GPU按需分配，不一定占用某个GPU的全部内存
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
#日志保存地址
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
timer = Timer()
for step in range(cfgs.MAX_ITERATION):
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, img_name, rpnLocLoss, rpnClsLoss, rpnTotalLoss, \
fastrcnnLocLoss, fastrcnnClsLoss, fastrcnnTotalLoss, totalLoss = \
sess.run(
[train_op, global_step, img_name_batch, rpn_location_loss, rpn_cls_loss, rpn_total_loss,
fastrcnn_loc_loss, fastrcnn_cls_loss, fastrcnn_total_loss, total_loss])
end = time.time()
print('{}: step:{}/{}iter'.format(training_time,
global_stepnp,
cfgs.MAX_ITERATION))
print('>>> rpn_loc_loss:{:.4f} | rpn_cla_loss:{:.4f} | rpn_total_loss:{:.4f}\n'
'>>> fast_rcnn_loc_loss:{:.4f} | fast_rcnn_cla_loss:{:.4f} | fast_rcnn_total_loss:{:.4f}\n'
'>>> single_time:{:.4f}s | total_loss:{:.4f} '\
.format(rpnLocLoss, rpnClsLoss,rpnTotalLoss, fastrcnnLocLoss, fastrcnnClsLoss,
fastrcnnTotalLoss,(end-start), totalLoss))
print(
'~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
)
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE
== 0) or (step == cfgs.MAX_ITERATION - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.mkdir(save_dir)
save_ckpt = os.path.join(
save_dir, 'voc_' + str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def eval_text(img_num, mode):
with tf.Graph().as_default():
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
# 通过readtfrecord来读取文件
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME,
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=False)
gtboxes_and_label = tf.py_func(
back_forward_convert,
inp=[tf.squeeze(gtboxes_and_label_batch, 0)],
Tout=tf.float32)
gtboxes_and_label = tf.reshape(gtboxes_and_label, [-1, 6])
gtboxes_and_label_minAreaRectangle = get_horizen_minAreaRectangle(
gtboxes_and_label)
gtboxes_and_label_minAreaRectangle = tf.reshape(
gtboxes_and_label_minAreaRectangle, [-1, 5])
final_boxes_h, final_scores_h, final_category_h, \
final_boxes_r, final_scores_r, final_category_r= faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch,
gtboxes_r_batch=gtboxes_and_label,
gtboxes_h_batch=gtboxes_and_label_minAreaRectangle)
if mode == 0:
final_boxes_r = get_horizen_minAreaRectangle(final_boxes_r, False)
# train
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
restorer, restore_ckpt = faster_rcnn.get_restorer()
config = tf.ConfigProto()
# config.gpu_options.per_process_gpu_memory_fraction = 0.5
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
gtboxes_horizontal_dict = {}
predict_horizontal_dict = {}
gtboxes_rotate_dict = {}
predict_rotate_dict = {}
for i in range(img_num):
start = time.time()
_img_name_batch, _img_batch, _gtboxes_and_label, _gtboxes_and_label_minAreaRectangle, \
_final_boxes_h, _final_scores_h, _final_category_h, _final_boxes_r, \
_final_scores_r, _final_category_r \
= sess.run([img_name_batch, img_batch, gtboxes_and_label, gtboxes_and_label_minAreaRectangle,
final_boxes_h, final_scores_h, final_category_h, final_boxes_r,
final_scores_r, final_category_r])
end = time.time()
# det_detections_h = draw_box_in_img.draw_box_cv(np.squeeze(_img_batch, 0),
# boxes=_final_boxes_h,
# labels=_final_category_h,
# scores=final_scores_h)
# det_detections_r = draw_box_in_img.draw_rotate_box_cv(np.squeeze(_img_batch, 0),
# boxes=_final_boxes_r,
# labels=_final_category_r,
# scores= _final_scores_r)
# save_dir = os.path.join('/home/sws/code/Tfrecord/', cfgs.VERSION)
# tools.mkdir(save_dir)
# cv2.imwrite(save_dir + '/' + _img_name_batch[0] + '_h.jpg',
# det_detections_h)
# cv2.imwrite(save_dir + '/' + _img_name_batch[0] + '_r.jpg',
# det_detections_r)
# gtboxes convert dict
gtboxes_horizontal_dict[str(_img_name_batch[0])] = []
predict_horizontal_dict[str(_img_name_batch[0])] = []
gtboxes_rotate_dict[str(_img_name_batch[0])] = []
predict_rotate_dict[str(_img_name_batch[0])] = []
gtbox_horizontal_list, predict_horizontal_list = \
make_dict_packle(_gtboxes_and_label_minAreaRectangle, _final_boxes_h,
_final_scores_h, _final_category_h)
# if mode == 0:
# gtbox_rotate_list, predict_rotate_list = \
# make_dict_packle(_gtboxes_and_label_minAreaRectangle, _fast_rcnn_decode_boxes_rotate,
# _fast_rcnn_score_rotate, _detection_category_rotate)
if mode == 0:
gtbox_rotate_list, predict_rotate_list = \
make_dict_packle(_gtboxes_and_label_minAreaRectangle, _final_boxes_r,
_final_scores_r, _final_category_r)
else:
gtbox_rotate_list, predict_rotate_list = \
make_dict_packle(_gtboxes_and_label, _final_boxes_r,
_final_scores_r, _final_category_r)
gtboxes_horizontal_dict[str(
_img_name_batch[0])].extend(gtbox_horizontal_list)
predict_horizontal_dict[str(
_img_name_batch[0])].extend(predict_horizontal_list)
gtboxes_rotate_dict[str(
_img_name_batch[0])].extend(gtbox_rotate_list)
predict_rotate_dict[str(
_img_name_batch[0])].extend(predict_rotate_list)
view_bar(
'{} image cost {}s'.format(str(_img_name_batch[0]),
(end - start)), i + 1, img_num)
print('\n')
fw1 = open('gtboxes_horizontal_dict.pkl', 'w')
fw2 = open('predict_horizontal_dict.pkl', 'w')
fw3 = open('gtboxes_rotate_dict.pkl', 'w')
fw4 = open('predict_rotate_dict.pkl', 'w')
pickle.dump(gtboxes_horizontal_dict, fw1)
pickle.dump(predict_horizontal_dict, fw2)
pickle.dump(gtboxes_rotate_dict, fw3)
pickle.dump(predict_rotate_dict, fw4)
fw1.close()
fw2.close()
fw3.close()
fw4.close()
coord.request_stop()
coord.join(threads)
def train():
with tf.Graph().as_default(), tf.device('/cpu:0'):
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(global_step,
boundaries=[np.int64(cfgs.DECAY_STEP[0]), np.int64(cfgs.DECAY_STEP[1]),
np.int64(cfgs.DECAY_STEP[2])],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100., cfgs.LR / 1000.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=True)
with tf.name_scope('get_batch'):
num_gpu = len(cfgs.GPU_GROUP.strip().split(','))
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch, img_h_batch, img_w_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE * num_gpu,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
# gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
# if cfgs.NET_NAME in ['resnet101_v1d', 'resnet50_v1d']:
# img_batch = img_batch / tf.constant([cfgs.PIXEL_STD])
# data processing
inputs_list = []
for i in range(num_gpu):
# img_name = img_name_batch[i]
img = tf.expand_dims(img_batch[i], axis=0)
gtboxes_and_label = tf.cast(tf.reshape(gtboxes_and_label_batch[i], [-1, 9]), tf.float32)
num_objects = num_objects_batch[i]
num_objects = tf.cast(tf.reshape(num_objects, [-1, ]), tf.float32)
img_h = img_h_batch[i]
img_w = img_w_batch[i]
# img_h = tf.cast(tf.reshape(img_h, [-1, ]), tf.float32)
# img_w = tf.cast(tf.reshape(img_w, [-1, ]), tf.float32)
inputs_list.append([img, gtboxes_and_label, num_objects, img_h, img_w])
tower_grads = []
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
total_loss_dict = {
'rpn_cls_loss': tf.constant(0., tf.float32),
'rpn_loc_loss': tf.constant(0., tf.float32),
'fastrcnn_cls_loss_h': tf.constant(0., tf.float32),
'fastrcnn_loc_loss_h': tf.constant(0., tf.float32),
'fastrcnn_cls_loss_r': tf.constant(0., tf.float32),
'fastrcnn_loc_loss_r': tf.constant(0., tf.float32),
'total_losses': tf.constant(0., tf.float32),
}
with tf.variable_scope(tf.get_variable_scope()):
for i in range(num_gpu):
with tf.device('/gpu:%d' % i):
with tf.name_scope('tower_%d' % i):
with slim.arg_scope(
[slim.model_variable, slim.variable],
device='/device:CPU:0'):
with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane,
slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
gtboxes_and_label = tf.py_func(get_gtboxes_and_label,
inp=[inputs_list[i][1], inputs_list[i][2]],
Tout=tf.float32)
gtboxes_and_label = tf.py_func(back_forward_convert,
inp=[gtboxes_and_label],
Tout=tf.float32)
gtboxes_and_label = tf.reshape(gtboxes_and_label, [-1, 6])
gtboxes_and_label_AreaRectangle = get_horizen_minAreaRectangle(gtboxes_and_label)
gtboxes_and_label_AreaRectangle = tf.reshape(gtboxes_and_label_AreaRectangle, [-1, 5])
img = inputs_list[i][0]
img_shape = inputs_list[i][-2:]
img = tf.image.crop_to_bounding_box(image=img,
offset_height=0,
offset_width=0,
target_height=tf.cast(img_shape[0], tf.int32),
target_width=tf.cast(img_shape[1], tf.int32))
final_boxes_h, final_scores_h, final_category_h, \
final_boxes_r, final_scores_r, final_category_r, loss_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img,
gtboxes_r_batch=gtboxes_and_label,
gtboxes_h_batch=gtboxes_and_label_AreaRectangle)
if cfgs.ADD_BOX_IN_TENSORBOARD:
dets_in_img = draw_boxes_with_categories_and_scores(img_batch=img,
boxes=final_boxes_h,
labels=final_category_h,
scores=final_scores_h)
dets_rotate_in_img = draw_boxes_with_categories_and_scores_rotate(img_batch=img,
boxes=final_boxes_r,
labels=final_category_r,
scores=final_scores_r)
tf.summary.image('Compare/final_detection_rotate_gpu:%d' % i, dets_rotate_in_img)
total_losses = 0.0
for k in loss_dict.keys():
total_losses += loss_dict[k]
total_loss_dict[k] += loss_dict[k] / num_gpu
total_losses = total_losses / num_gpu
total_loss_dict['total_losses'] += total_losses
if i == num_gpu - 1:
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
# weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
total_losses = total_losses + tf.add_n(regularization_losses)
tf.get_variable_scope().reuse_variables()
grads = optimizer.compute_gradients(total_losses) # average loss
tower_grads.append(grads)
for k in total_loss_dict.keys():
tf.summary.scalar('{}/{}'.format(k.split('_')[0], k), total_loss_dict[k])
if len(tower_grads) > 1:
grads = sum_gradients(tower_grads)
else:
grads = tower_grads[0]
apply_gradient_op = optimizer.apply_gradients(grads, global_step=global_step)
variable_averages = tf.train.ExponentialMovingAverage(0.9999, global_step)
variables_averages_op = variable_averages.apply(tf.trainable_variables())
train_op = tf.group(apply_gradient_op, variables_averages_op)
# train_op = optimizer.apply_gradients(final_gvs, global_step=global_step)
summary_op = tf.summary.merge_all()
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=10)
init_op = tf.group(
tf.global_variables_initializer(),
tf.local_variables_initializer()
)
tfconfig = tf.ConfigProto(
allow_soft_placement=True, log_device_placement=False)
#tfconfig.gpu_options.allow_growth = True
tfconfig.gpu_options.per_process_gpu_memory_fraction = 0.7
with tf.Session(config=tfconfig) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(coord=coord, sess=sess)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
for step in range(cfgs.MAX_ITERATION // num_gpu):
training_time = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, total_loss_dict_ = \
sess.run([train_op, global_step, total_loss_dict])
end = time.time()
print('***'*20)
print("""%s: global_step:%d current_step:%d"""
% (training_time, (global_stepnp-1)*num_gpu, step*num_gpu))
print("""per_cost_time:%.3fs"""
% ((end - start) / num_gpu))
loss_str = ''
for k in total_loss_dict_.keys():
loss_str += '%s:%.3f\n' % (k, total_loss_dict_[k])
print(loss_str)
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run([train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, (global_stepnp-1)*num_gpu)
summary_writer.flush()
if (step > 0 and step % (cfgs.SAVE_WEIGHTS_INTE // num_gpu) == 0) or (step >= cfgs.MAX_ITERATION // num_gpu - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.mkdir(save_dir)
save_ckpt = os.path.join(save_dir, 'coco_' + str((global_stepnp-1)*num_gpu) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def train():
with tf.Graph().as_default(), tf.device('/cpu:0'):
global_step = slim.get_or_create_global_step()
lr = warmup_lr(cfgs.LR, global_step, cfgs.WARM_SETP, cfgs.NUM_GPU*cfgs.BATCH_SIZE)
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=True,
batch_size=cfgs.BATCH_SIZE)
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch, img_h_batch, img_w_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE * cfgs.NUM_GPU,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
# data processing
inputs_list = []
for i in range(cfgs.NUM_GPU):
start = i*cfgs.BATCH_SIZE
end = (i+1)*cfgs.BATCH_SIZE
img = img_batch[start:end, :, :, :]
if cfgs.NET_NAME in ['resnet101_v1d', 'resnet50_v1d']:
img = img / tf.constant([cfgs.PIXEL_STD])
gtboxes_and_label = tf.cast(tf.reshape(gtboxes_and_label_batch[start:end, :, :],
[cfgs.BATCH_SIZE, -1, 5]), tf.float32)
num_objects = num_objects_batch[start:end]
num_objects = tf.cast(tf.reshape(num_objects, [cfgs.BATCH_SIZE, -1, ]), tf.float32)
img_h = img_h_batch[start:end]
img_w = img_w_batch[start:end]
# img_h = tf.cast(tf.reshape(img_h, [-1, ]), tf.float32)
# img_w = tf.cast(tf.reshape(img_w, [-1, ]), tf.float32)
inputs_list.append([img, gtboxes_and_label, num_objects, img_h, img_w])
# put_op_list = []
# get_op_list = []
# for i in range(cfgs.NUM_GPU):
# with tf.device("/GPU:%s" % i):
# area = tf.contrib.staging.StagingArea(
# dtypes=[tf.float32, tf.float32, tf.float32])
# put_op_list.append(area.put(inputs_list[i]))
# get_op_list.append(area.get())
tower_grads = []
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
total_loss_dict = {
'rpn_cls_loss': tf.constant(0., tf.float32),
'rpn_bbox_loss': tf.constant(0., tf.float32),
'rpn_ctr_loss': tf.constant(0., tf.float32),
'total_losses': tf.constant(0., tf.float32),
}
with tf.variable_scope(tf.get_variable_scope()):
for i in range(cfgs.NUM_GPU):
with tf.device('/gpu:%d' % i):
with tf.name_scope('tower_%d' % i):
with slim.arg_scope(
[slim.model_variable, slim.variable],
device='/device:CPU:0'):
with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane,
slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
gtboxes_and_label = tf.py_func(get_gtboxes_and_label,
inp=[inputs_list[i][1], inputs_list[i][2]],
Tout=tf.float32)
gtboxes_and_label = tf.reshape(gtboxes_and_label, [cfgs.BATCH_SIZE, -1, 5])
img = inputs_list[i][0]
img_shape = inputs_list[i][-2:]
h_crop = tf.reduce_max(img_shape[0])
w_crop = tf.reduce_max(img_shape[1])
img = tf.image.crop_to_bounding_box(image=img,
offset_height=0,
offset_width=0,
target_height=tf.cast(h_crop, tf.int32),
target_width=tf.cast(w_crop, tf.int32))
outputs = faster_rcnn.build_whole_detection_network(input_img_batch=img,
gtboxes_batch=gtboxes_and_label)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(img_batch=img[0, :, :, :],
boxes=gtboxes_and_label[0, :, :-1],
labels=gtboxes_and_label[0, :, -1])
gtboxes_in_img = tf.expand_dims(gtboxes_in_img, 0)
tf.summary.image('Compare/gtboxes_gpu:%d' % i, gtboxes_in_img)
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = show_box_in_tensor.draw_boxes_with_categories_and_scores(
img_batch=img[0, :, :, :],
boxes=outputs[0],
scores=outputs[1],
labels=outputs[2])
detections_in_img = tf.expand_dims(detections_in_img, 0)
tf.summary.image('Compare/final_detection_gpu:%d' % i, detections_in_img)
loss_dict = outputs[-1]
total_losses = 0.0
for k in loss_dict.keys():
total_losses += loss_dict[k]
total_loss_dict[k] += loss_dict[k] / cfgs.NUM_GPU
total_losses = total_losses / cfgs.NUM_GPU
total_loss_dict['total_losses'] += total_losses
if i == cfgs.NUM_GPU - 1:
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
# weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
total_losses = total_losses + tf.add_n(regularization_losses)
tf.get_variable_scope().reuse_variables()
grads = optimizer.compute_gradients(total_losses)
tower_grads.append(grads)
for k in total_loss_dict.keys():
tf.summary.scalar('{}/{}'.format(k.split('_')[0], k), total_loss_dict[k])
if len(tower_grads) > 1:
grads = sum_gradients(tower_grads)
else:
grads = tower_grads[0]
# final_gvs = []
# with tf.variable_scope('Gradient_Mult'):
# for grad, var in grads:
# scale = 1.
# # if '/biases:' in var.name:
# # scale *= 2.
# if 'conv_new' in var.name:
# scale *= 3.
# if not np.allclose(scale, 1.0):
# grad = tf.multiply(grad, scale)
# final_gvs.append((grad, var))
apply_gradient_op = optimizer.apply_gradients(grads, global_step=global_step)
variable_averages = tf.train.ExponentialMovingAverage(0.9999, global_step)
variables_averages_op = variable_averages.apply(tf.trainable_variables())
train_op = tf.group(apply_gradient_op, variables_averages_op)
# train_op = optimizer.apply_gradients(final_gvs, global_step=global_step)
summary_op = tf.summary.merge_all()
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=10)
init_op = tf.group(
tf.global_variables_initializer(),
tf.local_variables_initializer()
)
tfconfig = tf.ConfigProto(
allow_soft_placement=True, log_device_placement=False)
tfconfig.gpu_options.allow_growth = True
num_per_iter = cfgs.NUM_GPU * cfgs.BATCH_SIZE
with tf.Session(config=tfconfig) as sess:
sess.run(init_op)
# sess.run(tf.initialize_all_variables())
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(coord=coord, sess=sess)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
for step in range(cfgs.MAX_ITERATION // num_per_iter):
training_time = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, total_loss_dict_ = \
sess.run([train_op, global_step, total_loss_dict])
end = time.time()
print('***'*20)
print("""%s: global_step:%d current_step:%d"""
% (training_time, (global_stepnp-1)*num_per_iter, step*num_per_iter))
print("""per_cost_time:%.3fs"""
% ((end - start) / num_per_iter))
loss_str = ''
for k in total_loss_dict_.keys():
loss_str += '%s:%.3f\n' % (k, total_loss_dict_[k])
print(loss_str)
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run([train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, (global_stepnp-1)*num_per_iter)
summary_writer.flush()
if (step > 0 and step % (cfgs.SAVE_WEIGHTS_INTE // num_per_iter) == 0) or (step >= cfgs.MAX_ITERATION // cfgs.NUM_GPU - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.mkdir(save_dir)
save_ckpt = os.path.join(save_dir, 'coco_' + str((global_stepnp-1)*num_per_iter) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def train():
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
mask_batch = tf.py_func(
image_preprocess.get_mask,
[tf.squeeze(img_batch, 0),
tf.squeeze(gtboxes_and_label_batch, 0)], [tf.float32])
gtboxes_and_label = tf.py_func(
back_forward_convert,
inp=[tf.squeeze(gtboxes_and_label_batch, 0)],
Tout=tf.float32)
gtboxes_and_label = tf.reshape(gtboxes_and_label, [-1, 6])
gtboxes_and_label_AreaRectangle = get_horizen_minAreaRectangle(
gtboxes_and_label)
gtboxes_and_label_AreaRectangle = tf.reshape(
gtboxes_and_label_AreaRectangle, [-1, 5])
with tf.name_scope('draw_gtboxes'):
gtboxes_in_img = draw_box_with_color(
img_batch,
tf.reshape(gtboxes_and_label_AreaRectangle, [-1, 5])[:, :-1],
text=tf.shape(gtboxes_and_label_AreaRectangle)[0])
gtboxes_rotate_in_img = draw_box_with_color_rotate(
img_batch,
tf.reshape(gtboxes_and_label, [-1, 6])[:, :-1],
text=tf.shape(gtboxes_and_label)[0])
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
# list as many types of layers as possible, even if they are not used now
with slim.arg_scope([
slim.conv2d, slim.conv2d_in_plane, slim.conv2d_transpose,
slim.separable_conv2d, slim.fully_connected
],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
final_boxes_h, final_scores_h, final_category_h, \
final_boxes_r, final_scores_r, final_category_r, loss_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch,
gtboxes_r_batch=gtboxes_and_label,
gtboxes_h_batch=gtboxes_and_label_AreaRectangle,
mask_batch=mask_batch[0])
dets_in_img = draw_boxes_with_categories_and_scores(
img_batch=img_batch,
boxes=final_boxes_h,
labels=final_category_h,
scores=final_scores_h)
dets_rotate_in_img = draw_boxes_with_categories_and_scores_rotate(
img_batch=img_batch,
boxes=final_boxes_r,
labels=final_category_r,
scores=final_scores_r)
# ----------------------------------------------------------------------------------------------------build loss
weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
rpn_location_loss = loss_dict['rpn_loc_loss']
rpn_cls_loss = loss_dict['rpn_cls_loss']
rpn_total_loss = rpn_location_loss + rpn_cls_loss
fastrcnn_cls_loss_h = loss_dict['fastrcnn_cls_loss_h']
fastrcnn_loc_loss_h = loss_dict['fastrcnn_loc_loss_h']
fastrcnn_cls_loss_r = loss_dict['fastrcnn_cls_loss_r']
fastrcnn_loc_loss_r = loss_dict['fastrcnn_loc_loss_r']
fastrcnn_total_loss = fastrcnn_cls_loss_h + fastrcnn_loc_loss_h + fastrcnn_cls_loss_r + fastrcnn_loc_loss_r
if cfgs.ADD_ATTENTION:
attention_loss = loss_dict['attention_loss']
total_loss = rpn_total_loss + fastrcnn_total_loss + attention_loss + weight_decay_loss
else:
attention_loss = tf.convert_to_tensor(0)
total_loss = rpn_total_loss + fastrcnn_total_loss + weight_decay_loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RPN_LOSS/cls_loss', rpn_cls_loss)
tf.summary.scalar('RPN_LOSS/location_loss', rpn_location_loss)
tf.summary.scalar('RPN_LOSS/rpn_total_loss', rpn_total_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss_h', fastrcnn_cls_loss_h)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss_h',
fastrcnn_loc_loss_h)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss_r', fastrcnn_cls_loss_r)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss_r',
fastrcnn_loc_loss_r)
tf.summary.scalar('FAST_LOSS/fastrcnn_total_loss', fastrcnn_total_loss)
if cfgs.ADD_ATTENTION:
tf.summary.scalar('ATTENTION_LOSS/attention_loss', attention_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
tf.summary.image('img/gtboxes', gtboxes_in_img)
tf.summary.image('img/gtboxes_rotate', gtboxes_rotate_in_img)
tf.summary.image('img/mask', mask_batch)
tf.summary.image('img/dets', dets_in_img)
tf.summary.image('img/dets_rotate', dets_rotate_in_img)
# ---------------------------------------------------------------------------------------------add summary
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(
global_step,
boundaries=[
np.int64(cfgs.DECAY_STEP[0]),
np.int64(cfgs.DECAY_STEP[1])
],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
# ---------------------------------------------------------------------------------------------compute gradients
gradients = faster_rcnn.get_gradients(optimizer, total_loss)
# enlarge_gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = faster_rcnn.enlarge_gradients_for_bias(gradients)
if cfgs.GRADIENT_CLIPPING_BY_NORM:
with tf.name_scope('clip_gradients'):
gradients = slim.learning.clip_gradient_norms(
gradients, cfgs.GRADIENT_CLIPPING_BY_NORM)
# ---------------------------------------------------------------------------------------------compute gradients
# train_op
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
summary_op = tf.summary.merge_all()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=10)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
#
with tf.Session(config=config) as sess:
# with tf.Session() as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
for step in range(cfgs.MAX_ITERATION):
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_global_step, _img_name_batch, _rpn_location_loss, _rpn_classification_loss, \
_rpn_total_loss, _fast_rcnn_location_loss, _fast_rcnn_classification_loss, \
_fast_rcnn_location_rotate_loss, _fast_rcnn_classification_rotate_loss, \
_fast_rcnn_total_loss, _attention_loss, _total_loss, _ = \
sess.run([global_step, img_name_batch, rpn_location_loss, rpn_cls_loss,
rpn_total_loss, fastrcnn_loc_loss_h, fastrcnn_cls_loss_h,
fastrcnn_loc_loss_r, fastrcnn_cls_loss_r, fastrcnn_total_loss,
attention_loss, total_loss, train_op])
end = time.time()
print(""" {}: step{} image_name:{} |\t
rpn_loc_loss:{} |\t rpn_cla_loss:{} |\t
rpn_total_loss:{} |
fast_rcnn_loc_loss:{} |\t fast_rcnn_cla_loss:{} |\t
fast_rcnn_loc_rotate_loss:{} |\t fast_rcnn_cla_rotate_loss:{} |\t
fast_rcnn_total_loss:{} |\t attention_loss:{} |\t
total_loss:{} |\t pre_cost_time:{}s""" \
.format(training_time, _global_step, str(_img_name_batch[0]), _rpn_location_loss,
_rpn_classification_loss, _rpn_total_loss, _fast_rcnn_location_loss,
_fast_rcnn_classification_loss, _fast_rcnn_location_rotate_loss,
_fast_rcnn_classification_rotate_loss, _fast_rcnn_total_loss,
_attention_loss, _total_loss, (end - start)))
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE
== 0) or (step == cfgs.MAX_ITERATION - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
save_ckpt = os.path.join(
save_dir, 'voc_' + str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def train():
retinanet = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
if cfgs.NET_NAME in ['resnet152_v1d', 'resnet101_v1d', 'resnet50_v1d']:
img_batch = img_batch / tf.constant([cfgs.PIXEL_STD])
final_bbox, final_scores, final_category, loss_dict = retinanet.build_whole_detection_network(
input_img_batch=img_batch, gtboxes_batch=gtboxes_and_label)
# ----------------------------------------------------------------------------------------------------build loss
weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
cls_loss = loss_dict['cls_loss']
reg_loss = loss_dict['reg_loss']
total_loss = cls_loss + reg_loss + weight_decay_loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RETINANET_LOSS/cls_loss', cls_loss)
tf.summary.scalar('RETINANET_LOSS/reg_loss', reg_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(
img_batch=img_batch,
boxes=gtboxes_and_label[:, :-1],
labels=gtboxes_and_label[:, -1])
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = show_box_in_tensor.draw_boxes_with_categories_and_scores(
img_batch=img_batch,
boxes=final_bbox,
labels=final_category,
scores=final_scores)
tf.summary.image('Compare/final_detection', detections_in_img)
tf.summary.image('Compare/gtboxes', gtboxes_in_img)
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(
global_step,
boundaries=[
np.int64(cfgs.DECAY_STEP[0]),
np.int64(cfgs.DECAY_STEP[1])
],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
# ---------------------------------------------------------------------------------------------compute gradients
gradients = retinanet.get_gradients(optimizer, total_loss)
# enlarge_gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = retinanet.enlarge_gradients_for_bias(gradients)
if cfgs.GRADIENT_CLIPPING_BY_NORM:
with tf.name_scope('clip_gradients'):
gradients = slim.learning.clip_gradient_norms(
gradients, cfgs.GRADIENT_CLIPPING_BY_NORM)
# train_op
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
summary_op = tf.summary.merge_all()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
restorer, restore_ckpt = retinanet.get_restorer()
saver = tf.train.Saver(max_to_keep=30)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
if not os.path.exists(summary_path):
os.makedirs(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
for step in range(cfgs.MAX_ITERATION):
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, img_name, reg_loss_, cls_loss_, total_loss_ = \
sess.run(
[train_op, global_step, img_name_batch, reg_loss, cls_loss, total_loss])
end = time.time()
print(""" {}: step{} image_name:{} |\t
reg_loss:{} |\t cls_loss:{} |\t total_loss:{} |per_cost_time:{}s""" \
.format(training_time, global_stepnp, str(img_name[0]), reg_loss_, cls_loss_, total_loss_,
(end - start)))
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE
== 0) or (step == cfgs.MAX_ITERATION - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
save_ckpt = os.path.join(
save_dir, '{}_'.format(cfgs.DATASET_NAME) +
str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def train():
faster_rcnn = build_whole_network.DetectionNetwork(base_network_name=cfgs.NET_NAME,
is_training=True)
with tf.name_scope('get_batch'):
img_plac = tf.placeholder(dtype=tf.uint8, shape=[None, None, 3])
gtbox_plac = tf.placeholder(dtype=tf.int32, shape=[None, 5])
img_batch, gtboxes_and_label = preprocess_img(img_plac, gtbox_plac)
# gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
# list as many types of layers as possible, even if they are not used now
with slim.arg_scope([slim.conv2d, slim.conv2d_in_plane, \
slim.conv2d_transpose, slim.separable_conv2d, slim.fully_connected],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
final_bbox, final_scores, final_category, loss_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch,
gtboxes_batch=gtboxes_and_label)
# ----------------------------------------------------------------------------------------------------build loss
weight_decay_loss = 0 # tf.add_n(slim.losses.get_regularization_losses())
rpn_location_loss = loss_dict['rpn_loc_loss']
rpn_cls_loss = loss_dict['rpn_cls_loss']
rpn_total_loss = rpn_location_loss + rpn_cls_loss
fastrcnn_cls_loss = loss_dict['fastrcnn_cls_loss']
fastrcnn_loc_loss = loss_dict['fastrcnn_loc_loss']
fastrcnn_total_loss = fastrcnn_cls_loss + fastrcnn_loc_loss
total_loss = rpn_total_loss + fastrcnn_total_loss + weight_decay_loss
# ____________________________________________________________________________________________________build loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RPN_LOSS/cls_loss', rpn_cls_loss)
tf.summary.scalar('RPN_LOSS/location_loss', rpn_location_loss)
tf.summary.scalar('RPN_LOSS/rpn_total_loss', rpn_total_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss', fastrcnn_cls_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss', fastrcnn_loc_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_total_loss', fastrcnn_total_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(img_batch=img_batch,
boxes=gtboxes_and_label[:, :-1],
labels=gtboxes_and_label[:, -1])
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = show_box_in_tensor.draw_boxes_with_categories_and_scores(img_batch=img_batch,
boxes=final_bbox,
labels=final_category,
scores=final_scores)
tf.summary.image('Compare/final_detection', detections_in_img)
tf.summary.image('Compare/gtboxes', gtboxes_in_img)
# ___________________________________________________________________________________________________add summary
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(global_step,
boundaries=[np.int64(cfgs.DECAY_STEP[0]), np.int64(cfgs.DECAY_STEP[1])],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
# ---------------------------------------------------------------------------------------------compute gradients
gradients = faster_rcnn.get_gradients(optimizer, total_loss)
# enlarge_gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = faster_rcnn.enlarge_gradients_for_bias(gradients)
if cfgs.GRADIENT_CLIPPING_BY_NORM:
with tf.name_scope('clip_gradients_YJR'):
gradients = slim.learning.clip_gradient_norms(gradients,
cfgs.GRADIENT_CLIPPING_BY_NORM)
# _____________________________________________________________________________________________compute gradients
# train_op
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
summary_op = tf.summary.merge_all()
init_op = tf.group(
tf.global_variables_initializer(),
tf.local_variables_initializer()
)
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=30)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
for step in range(cfgs.MAX_ITERATION):
img_id, img, gt_info = next_img(step=step)
training_time = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step],
feed_dict={img_plac: img,
gtbox_plac: gt_info}
)
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, rpnLocLoss, rpnClsLoss, rpnTotalLoss, \
fastrcnnLocLoss, fastrcnnClsLoss, fastrcnnTotalLoss, totalLoss = \
sess.run(
[train_op, global_step, rpn_location_loss, rpn_cls_loss, rpn_total_loss,
fastrcnn_loc_loss, fastrcnn_cls_loss, fastrcnn_total_loss, total_loss],
feed_dict={img_plac: img,
gtbox_plac: gt_info})
end = time.time()
print(""" {}: step{} image_name:{} |\t
rpn_loc_loss:{} |\t rpn_cla_loss:{} |\t rpn_total_loss:{} |
fast_rcnn_loc_loss:{} |\t fast_rcnn_cla_loss:{} |\t fast_rcnn_total_loss:{} |
total_loss:{} |\t per_cost_time:{}s""" \
.format(training_time, global_stepnp, str(img_id), rpnLocLoss, rpnClsLoss,
rpnTotalLoss, fastrcnnLocLoss, fastrcnnClsLoss, fastrcnnTotalLoss, totalLoss,
(end - start)))
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run([train_op, global_step, summary_op],
feed_dict={img_plac: img,
gtbox_plac: gt_info}
)
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE == 0) or (step == cfgs.MAX_ITERATION - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.mkdir(save_dir)
save_ckpt = os.path.join(save_dir, 'voc_' + str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
def train():
with tf.Graph().as_default(), tf.device('/cpu:0'):
num_gpu = len(cfgs.GPU_GROUP.strip().split(','))
global_step = slim.get_or_create_global_step()
lr = warmup_lr(cfgs.LR, global_step, cfgs.WARM_SETP, num_gpu)
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
retinanet = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope('get_batch'):
if cfgs.IMAGE_PYRAMID:
shortside_len_list = tf.constant(cfgs.IMG_SHORT_SIDE_LEN)
shortside_len = tf.random_shuffle(shortside_len_list)[0]
else:
shortside_len = cfgs.IMG_SHORT_SIDE_LEN
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch, img_h_batch, img_w_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME,
batch_size=cfgs.BATCH_SIZE * num_gpu,
shortside_len=shortside_len,
is_training=True)
# data processing
inputs_list = []
for i in range(num_gpu):
img = tf.expand_dims(img_batch[i], axis=0)
if cfgs.NET_NAME in [
'resnet152_v1d', 'resnet101_v1d', 'resnet50_v1d'
]:
img = img / tf.constant([cfgs.PIXEL_STD])
gtboxes_and_label_r = tf.py_func(backward_convert,
inp=[gtboxes_and_label_batch[i]],
Tout=tf.float32)
gtboxes_and_label_r = tf.reshape(gtboxes_and_label_r, [-1, 6])
gtboxes_and_label_h = get_horizen_minAreaRectangle(
gtboxes_and_label_batch[i])
gtboxes_and_label_h = tf.reshape(gtboxes_and_label_h, [-1, 5])
num_objects = num_objects_batch[i]
num_objects = tf.cast(tf.reshape(num_objects, [
-1,
]), tf.float32)
img_h = img_h_batch[i]
img_w = img_w_batch[i]
inputs_list.append([
img, gtboxes_and_label_h, gtboxes_and_label_r, num_objects,
img_h, img_w
])
tower_grads = []
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(
cfgs.WEIGHT_DECAY)
total_loss_dict = {
'cls_loss': tf.constant(0., tf.float32),
'reg_loss': tf.constant(0., tf.float32),
'total_losses': tf.constant(0., tf.float32),
}
with tf.variable_scope(tf.get_variable_scope()):
for i in range(num_gpu):
with tf.device('/gpu:%d' % i):
with tf.name_scope('tower_%d' % i):
with slim.arg_scope(
[slim.model_variable, slim.variable],
device='/device:CPU:0'):
with slim.arg_scope(
[
slim.conv2d, slim.conv2d_in_plane,
slim.conv2d_transpose,
slim.separable_conv2d, slim.fully_connected
],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(
0.0)):
gtboxes_and_label_h, gtboxes_and_label_r = tf.py_func(
get_gtboxes_and_label,
inp=[
inputs_list[i][1], inputs_list[i][2],
inputs_list[i][3]
],
Tout=[tf.float32, tf.float32])
gtboxes_and_label_h = tf.reshape(
gtboxes_and_label_h, [-1, 5])
gtboxes_and_label_r = tf.reshape(
gtboxes_and_label_r, [-1, 6])
img = inputs_list[i][0]
img_shape = inputs_list[i][-2:]
img = tf.image.crop_to_bounding_box(
image=img,
offset_height=0,
offset_width=0,
target_height=tf.cast(
img_shape[0], tf.int32),
target_width=tf.cast(
img_shape[1], tf.int32))
outputs = retinanet.build_whole_detection_network(
input_img_batch=img,
gtboxes_batch_h=gtboxes_and_label_h,
gtboxes_batch_r=gtboxes_and_label_r,
gpu_id=i)
gtboxes_in_img_h = draw_boxes_with_categories(
img_batch=img,
boxes=gtboxes_and_label_h[:, :-1],
labels=gtboxes_and_label_h[:, -1],
method=0)
gtboxes_in_img_r = draw_boxes_with_categories(
img_batch=img,
boxes=gtboxes_and_label_r[:, :-1],
labels=gtboxes_and_label_r[:, -1],
method=1)
tf.summary.image(
'Compare/gtboxes_h_gpu:%d' % i,
gtboxes_in_img_h)
tf.summary.image(
'Compare/gtboxes_r_gpu:%d' % i,
gtboxes_in_img_r)
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = draw_boxes_with_categories_and_scores(
img_batch=img,
boxes=outputs[0],
scores=outputs[1],
labels=outputs[2],
method=1)
tf.summary.image(
'Compare/final_detection_gpu:%d' % i,
detections_in_img)
loss_dict = outputs[-1]
total_losses = 0.0
for k in loss_dict.keys():
total_losses += loss_dict[k]
total_loss_dict[
k] += loss_dict[k] / num_gpu
total_losses /= num_gpu
total_loss_dict['total_losses'] += total_losses
if i == num_gpu - 1:
regularization_losses = tf.get_collection(
tf.GraphKeys.REGULARIZATION_LOSSES)
# weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
total_losses = total_losses + tf.add_n(
regularization_losses)
tf.get_variable_scope().reuse_variables()
grads = optimizer.compute_gradients(total_losses)
if cfgs.GRADIENT_CLIPPING_BY_NORM is not None:
grads = slim.learning.clip_gradient_norms(
grads, cfgs.GRADIENT_CLIPPING_BY_NORM)
tower_grads.append(grads)
for k in total_loss_dict.keys():
tf.summary.scalar('{}/{}'.format(k.split('_')[0], k),
total_loss_dict[k])
if len(tower_grads) > 1:
grads = sum_gradients(tower_grads)
else:
grads = tower_grads[0]
if cfgs.MUTILPY_BIAS_GRADIENT is not None:
final_gvs = []
with tf.variable_scope('Gradient_Mult'):
for grad, var in grads:
scale = 1.
if '/biases:' in var.name:
scale *= cfgs.MUTILPY_BIAS_GRADIENT
if 'conv_new' in var.name:
scale *= 3.
if not np.allclose(scale, 1.0):
grad = tf.multiply(grad, scale)
final_gvs.append((grad, var))
apply_gradient_op = optimizer.apply_gradients(
final_gvs, global_step=global_step)
else:
apply_gradient_op = optimizer.apply_gradients(
grads, global_step=global_step)
variable_averages = tf.train.ExponentialMovingAverage(
0.9999, global_step)
variables_averages_op = variable_averages.apply(
tf.trainable_variables())
train_op = tf.group(apply_gradient_op, variables_averages_op)
# train_op = optimizer.apply_gradients(final_gvs, global_step=global_step)
summary_op = tf.summary.merge_all()
restorer, restore_ckpt = retinanet.get_restorer()
saver = tf.train.Saver(max_to_keep=5)
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
tfconfig = tf.ConfigProto(allow_soft_placement=True,
log_device_placement=False)
tfconfig.gpu_options.allow_growth = True
with tf.Session(config=tfconfig) as sess:
sess.run(init_op)
# sess.run(tf.initialize_all_variables())
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(coord=coord, sess=sess)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path,
graph=sess.graph)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
for step in range(cfgs.MAX_ITERATION // num_gpu):
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, total_loss_dict_ = \
sess.run([train_op, global_step, total_loss_dict])
end = time.time()
print('***' * 20)
print("""%s: global_step:%d current_step:%d""" %
(training_time,
(global_stepnp - 1) * num_gpu, step * num_gpu))
print("""per_cost_time:%.3fs""" %
((end - start) / num_gpu))
loss_str = ''
for k in total_loss_dict_.keys():
loss_str += '%s:%.3f\n' % (k, total_loss_dict_[k])
print(loss_str)
if np.isnan(total_loss_dict_['total_losses']):
sys.exit(0)
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
summary_writer.add_summary(
summary_str, (global_stepnp - 1) * num_gpu)
summary_writer.flush()
if (step > 0 and step % (cfgs.SAVE_WEIGHTS_INTE // num_gpu)
== 0) or (step >= cfgs.MAX_ITERATION // num_gpu - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.mkdir(save_dir)
save_ckpt = os.path.join(
save_dir, '{}_'.format(cfgs.DATASET_NAME) + str(
(global_stepnp - 1) * num_gpu) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def train():
# get Network Class
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
# get batch
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
# convert gtboxes_and_labels
gtboxes_and_label = tf.py_func(
back_forward_convert,
inp=[tf.squeeze(gtboxes_and_label_batch, 0)],
Tout=tf.float32)
gtboxes_and_label = tf.reshape(gtboxes_and_label, [-1, 6])
# draw ground-truth
with tf.name_scope('draw_gtboxes'):
gtboxes_in_img = draw_box_with_color_rotate(
img_batch,
tf.reshape(gtboxes_and_label, [-1, 6])[:, :-1],
text=tf.shape(gtboxes_and_label)[0])
# default regularizers
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
# list as many types of layers as possible, even if they are not used now
with slim.arg_scope([
slim.conv2d, slim.conv2d_in_plane, slim.conv2d_transpose,
slim.separable_conv2d, slim.fully_connected
],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
# build network
final_boxes, final_scores, final_category, loss_dict = \
faster_rcnn.build_whole_detection_network(input_img_batch=img_batch,
gtboxes_batch=gtboxes_and_label)
# draw detections
dets_in_img = draw_boxes_with_categories_and_scores_rotate(
img_batch=img_batch,
boxes=final_boxes,
labels=final_category,
scores=final_scores)
# ----------------------------------------------------------------------------------------------------build loss
# weight decay loss
weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
# rpn losses
rpn_location_loss = loss_dict['rpn_loc_loss']
rpn_cls_loss = loss_dict['rpn_cls_loss']
rpn_total_loss = rpn_location_loss + rpn_cls_loss
# fastrcnn losses
fastrcnn_cls_loss = loss_dict['fastrcnn_cls_loss']
fastrcnn_loc_loss = loss_dict['fastrcnn_loc_loss']
fastrcnn_total_loss = fastrcnn_cls_loss + fastrcnn_loc_loss
# total loss
total_loss = rpn_total_loss + fastrcnn_total_loss + weight_decay_loss
# ____________________________________________________________________________________________________build loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RPN_LOSS/cls_loss', rpn_cls_loss)
tf.summary.scalar('RPN_LOSS/location_loss', rpn_location_loss)
tf.summary.scalar('RPN_LOSS/rpn_total_loss', rpn_total_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss', fastrcnn_cls_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss', fastrcnn_loc_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_total_loss', fastrcnn_total_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
tf.summary.image('img/gtboxes', gtboxes_in_img)
tf.summary.image('img/dets', dets_in_img)
# ___________________________________________________________________________________________________add summary
# create global step
global_step = slim.get_or_create_global_step()
# learning rate setting
lr = tf.train.piecewise_constant(
global_step,
boundaries=[
np.int64(cfgs.DECAY_STEP[0]),
np.int64(cfgs.DECAY_STEP[1])
],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
# optimizer with lr
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
# ---------------------------------------------------------------------------------------------compute gradients
# compute gradients
gradients = faster_rcnn.get_gradients(optimizer, total_loss)
# enlarge gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = faster_rcnn.enlarge_gradients_for_bias(gradients)
# clipping gradients
if cfgs.GRADIENT_CLIPPING_BY_NORM:
with tf.name_scope('clip_gradients'):
gradients = slim.learning.clip_gradient_norms(
gradients, cfgs.GRADIENT_CLIPPING_BY_NORM)
# _____________________________________________________________________________________________compute gradients
# train_op, ie. apply gradients
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
# summary_op
summary_op = tf.summary.merge_all()
# init_op
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
# restorer and Saver
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=10)
# GPU Config
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
# start a session
with tf.Session(config=config) as sess:
# init
sess.run(init_op)
# restore
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
# Session对象是支持多线程的, 可以在同一个会话（Session）中创建多个线程，并行执行;
# 创建线程协调器(管理器), 并启动(Tensor/训练数据)入队线程,
# 入队具体使用多少个线程在read_tfrecord.py->next_batch->tf.train.batch中定义;
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
# construct summary writer
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
tools.mkdir(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
# train MAX_ITERATION steps
for step in range(cfgs.MAX_ITERATION):
# time of this step
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
# no show & no summary
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
# global_step is added to run, global_stepnp will be used in Save
_, global_stepnp = sess.run([train_op, global_step])
else:
# show
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
# middle infos are added to run, losses and img_name_batch
_global_step, _img_name_batch, _rpn_location_loss, _rpn_classification_loss, \
_rpn_total_loss, _fast_rcnn_location_loss, _fast_rcnn_classification_loss, \
_fast_rcnn_total_loss, _total_loss, _ = \
sess.run([global_step, img_name_batch, rpn_location_loss, rpn_cls_loss,
rpn_total_loss, fastrcnn_loc_loss, fastrcnn_cls_loss,
fastrcnn_total_loss, total_loss, train_op])
# # show boxes, scores, category infos
# final_boxes_r, _final_scores_r, _final_category_r = sess.run([final_boxes_r, final_scores_r, final_category_r])
# print('*'*100)
# print(_final_boxes_r)
# print(_final_scores_r)
# print(_final_category_r)
end = time.time()
# print
print(""" {}: step {}: image_name:{} |\t
rpn_loc_loss:{} |\t rpn_cla_loss:{} |\t
rpn_total_loss:{} |
fast_rcnn_loc_loss:{} |\t fast_rcnn_cla_loss:{} |\t
fast_rcnn_total_loss:{} |\t
total_loss:{} |\t pre_cost_time:{}s""" \
.format(training_time, _global_step, str(_img_name_batch[0]),
_rpn_location_loss, _rpn_classification_loss, _rpn_total_loss,
_fast_rcnn_location_loss, _fast_rcnn_classification_loss, _fast_rcnn_total_loss,
_total_loss, (end - start)))
# summary
else:
if step % cfgs.SMRY_ITER == 0:
# summary_op is added to run
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
# add summary
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
# save
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE
== 0) or (step == cfgs.MAX_ITERATION - 1):
# mkdir
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
# save checkpoint
save_ckpt = os.path.join(
save_dir, 'voc_' + str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
# 协调器发出终止所有线程的命令
coord.request_stop()
# 把开启的线程加入主线程，等待threads结束
coord.join(threads)
def train():
# Step 1:
# clw note：传递网络名称如resnet_v1，是否训练is_training，以及每个位置含有anchor box的个数，
# 构建基本的网络
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope(
'get_batch'
): # clw note：tf.name_scope 主要结合 tf.Variable() 来使用，方便参数命名管理。
# clw note：从文件队列、内存队列中读取、组合，得到该batch的内容
# 主要包括每个批次（目前仅支持批次数目即batch_size=1，也就是这里每次只读出1张图片）
# 对应的变量包括：图片名称、图片矩阵、ground truth坐标及对应的label，图片中包含的目标数
# 这些变量的组成结构均为 [批次数目，相应批次中每一幅图片的相关信息]
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
# clw note：样本个数m不知道，但是对单个样本都有gtboxes的4个坐标，加上1个label共5个值；使用-1来自动计算样本个数
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
# Step 2：
# clw note：Faster R-CNN网络的搭建！
# 先看一下下面这个函数arg_scope的声明
# @tf_contextlib.contextmanager
# def arg_scope(list_ops_or_scope, **kwargs): 功能是给list_ops中的内容设置默认值，即list中所有元素都用**kargs的参数设置。
# 有函数修饰符@tf_contextlib.contextmanager修饰arg_scope函数：@之后一般接一个可调用对象为其执行一系列辅助操作，
# 我们来看一个demo：
#########################################
# import time
# def my_time(func):
# print(time.ctime())
# return func()
#
# @my_time # 从这里可以看出@time 等价于 time(xxx()),但是这种写法你得考虑python代码的执行顺序
# def xxx():
# print('Hello world!')
#
# 运行结果：
# Wed Jul 26 23:01:21 2017
# Hello world!
##########################################
# 在这个例子中，xxx函数实现我们的主要功能，打印Hello world，但我们想给xxx函数添加一些辅助操作，让它同时打印出时间，于是我们用
# 函数修饰符 @ my_time完成这个目标。整个例子的执行流程为调用my_time可调用对象，它接受xxx函数作为参数，先打印时间，再执行xxx函数
# 详见：https://www.cnblogs.com/zzy-tf/p/9356883.html
# 来看另一个demo：
##########################################
# with slim.arg_scope(
# [slim.conv2d, slim.max_pool2d, slim.avg_pool2d],stride = 1, padding = 'VALID'):
# net = slim.conv2d(inputs, 32, [3, 3], stride = 2, scope = 'Conv2d_1a_3x3')
# net = slim.conv2d(net, 32, [3, 3], scope = 'Conv2d_2a_3x3')
# net = slim.conv2d(net, 64, [3, 3], padding = 'SAME', scope = 'Conv2d_2b_3x3')
# 所以，在使用过程中可以直接slim.conv2d( )等函数设置默认参数。例如在下面的代码中，不做单独声明的情况下，
# slim.conv2d, slim.max_pool2d, slim.avg_pool2d三个函数默认的步长都设为1，padding模式都是'VALID'的。
# 当然也可以在调用时进行单独声明，只不过一个一个写很麻烦，不如统一给个默认值。
# 这种参数设置方式在构建网络模型时，尤其是较深的网络时，可以节省时间。
with slim.arg_scope([
slim.conv2d, slim.conv2d_in_plane, slim.conv2d_transpose,
slim.separable_conv2d, slim.fully_connected
],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(
0.0)): # list as many types of layers as possible,
# even if they are not used now
# build_whole_detection_network功能：构建整体网络架构，包含backbone，RPN网络，Pooling层，以及后续网络。
# return：网络的最后的预测框，预测的类别信息，预测的概率，以及整体网络和RPN网络的损失，所有的损失被写入到一个字典中。
final_bbox, final_scores, final_category, loss_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch, gtboxes_batch=gtboxes_and_label)
# ----------------------------------------------------------------------------------------------------build loss
# weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
# weight_decay_loss = tf.add_n(tf.losses.get_regularization_losses())
rpn_location_loss = loss_dict['rpn_loc_loss']
rpn_cls_loss = loss_dict['rpn_cls_loss']
rpn_total_loss = rpn_location_loss + rpn_cls_loss
fastrcnn_cls_loss = loss_dict['fastrcnn_cls_loss']
fastrcnn_loc_loss = loss_dict['fastrcnn_loc_loss']
fastrcnn_total_loss = fastrcnn_cls_loss + fastrcnn_loc_loss
# clw note：根据论文的公式，最后将RPN网络的（分类，回归）误差与Fast-RCNN的(分类，回归）误差相加后作为总的误差进行训练即可。
total_loss = rpn_total_loss + fastrcnn_total_loss
# ____________________________________________________________________________________________________build loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RPN_LOSS/cls_loss', rpn_cls_loss)
tf.summary.scalar('RPN_LOSS/location_loss', rpn_location_loss)
tf.summary.scalar('RPN_LOSS/rpn_total_loss', rpn_total_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss', fastrcnn_cls_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss', fastrcnn_loc_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_total_loss', fastrcnn_total_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
# tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(
img_batch=img_batch,
boxes=gtboxes_and_label[:, :-1],
labels=gtboxes_and_label[:, -1])
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = show_box_in_tensor.draw_boxes_with_categories_and_scores(
img_batch=img_batch,
boxes=final_bbox,
labels=final_category,
scores=final_scores)
tf.summary.image('Compare/final_detection', detections_in_img)
tf.summary.image('Compare/gtboxes', gtboxes_in_img)
# ___________________________________________________________________________________________________add summary
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(
global_step,
boundaries=[
np.int64(cfgs.DECAY_STEP[0]),
np.int64(cfgs.DECAY_STEP[1])
],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(
lr, momentum=cfgs.MOMENTUM) # clw note：选择优化器，可以尝试其他选择，
# 也可以尝试tf.train.AdamOptimizer(1e-4).minimize(total_loss)
# ---------------------------------------------------------------------------------------------compute gradients
# clw note：对于上面优化器没有使用minimize()的几点说明，
# 使用minimize()操作，该操作不仅可以计算出梯度，而且还可以将梯度作用在变量上。
# 如果想按照自己的方式处理梯度，可以按照以下步骤：
# 1、使用compute_gradients()计算梯度，其实下面的get_gradients()方法就是optimizer.compute_gradients(loss)
# 2、使用自己的方式进一步处理梯度
# 3、使用apply_gradients()应用处理过后的梯度；
gradients = faster_rcnn.get_gradients(optimizer, total_loss)
# enlarge_gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = faster_rcnn.enlarge_gradients_for_bias(gradients)
if cfgs.GRADIENT_CLIPPING_BY_NORM: # clw note：clip_by_norm是指对梯度进行裁剪，通过控制梯度的最大范式，防止梯度爆炸的问题，是一种比较常用的梯度规约的方式
with tf.name_scope('clip_gradients_YJR'):
gradients = slim.learning.clip_gradient_norms(
gradients, cfgs.GRADIENT_CLIPPING_BY_NORM)
# _____________________________________________________________________________________________compute gradients
# train_op
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
summary_op = tf.summary.merge_all()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=30)
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
# tools.mkdir(summary_path)
if not os.path.exists(summary_path):
os.makedirs(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
for step in range(cfgs.MAX_ITERATION):
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, img_name, rpnLocLoss, rpnClsLoss, rpnTotalLoss, \
fastrcnnLocLoss, fastrcnnClsLoss, fastrcnnTotalLoss, totalLoss = \
sess.run(
[train_op, global_step, img_name_batch, rpn_location_loss, rpn_cls_loss, rpn_total_loss,
fastrcnn_loc_loss, fastrcnn_cls_loss, fastrcnn_total_loss, total_loss])
end = time.time()
print(""" {}: step{} image_name:{} |\t
rpn_loc_loss:{} |\t rpn_cla_loss:{} |\t rpn_total_loss:{} |
fast_rcnn_loc_loss:{} |\t fast_rcnn_cla_loss:{} |\t fast_rcnn_total_loss:{} |
total_loss:{} |\t per_cost_time:{}s""" \
.format(training_time, global_stepnp, str(img_name[0]), rpnLocLoss, rpnClsLoss,
rpnTotalLoss, fastrcnnLocLoss, fastrcnnClsLoss, fastrcnnTotalLoss, totalLoss,
(end - start)))
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE
== 0) or (step == cfgs.MAX_ITERATION - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
save_ckpt = os.path.join(
save_dir, 'voc_' + str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)
def tower_loss(scope):
with tf.name_scope(scope):
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(
cfgs.WEIGHT_DECAY)
# list as many types of layers as possible, even if they are not used now
with slim.arg_scope([
slim.conv2d, slim.conv2d_in_plane, slim.conv2d_transpose,
slim.separable_conv2d, slim.fully_connected
],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
final_bbox, final_scores, final_category, loss_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch, gtboxes_batch=gtboxes_and_label)
# ----------------------------------------------------------------------------------------------------build loss
weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
rpn_location_loss = loss_dict['rpn_loc_loss']
rpn_cls_loss = loss_dict['rpn_cls_loss']
rpn_total_loss = rpn_location_loss + rpn_cls_loss
fastrcnn_cls_loss = loss_dict['fastrcnn_cls_loss']
fastrcnn_loc_loss = loss_dict['fastrcnn_loc_loss']
fastrcnn_total_loss = fastrcnn_cls_loss + fastrcnn_loc_loss
total_loss = rpn_total_loss + fastrcnn_total_loss + weight_decay_loss
# ____________________________________________________________________________________________________build loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('RPN_LOSS/cls_loss', rpn_cls_loss)
tf.summary.scalar('RPN_LOSS/location_loss', rpn_location_loss)
tf.summary.scalar('RPN_LOSS/rpn_total_loss', rpn_total_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_cls_loss', fastrcnn_cls_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_location_loss',
fastrcnn_loc_loss)
tf.summary.scalar('FAST_LOSS/fastrcnn_total_loss', fastrcnn_total_loss)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(
img_batch=img_batch,
boxes=gtboxes_and_label[:, :-1],
labels=gtboxes_and_label[:, -1])
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img = show_box_in_tensor.draw_boxes_with_categories_and_scores(
img_batch=img_batch,
boxes=final_bbox,
labels=final_category,
scores=final_scores)
tf.summary.image('Compare/final_detection', detections_in_img)
tf.summary.image('Compare/gtboxes', gtboxes_in_img)
return total_loss, faster_rcnn, img_name_batch, rpn_location_loss, rpn_cls_loss, rpn_total_loss,\
fastrcnn_loc_loss, fastrcnn_cls_loss, fastrcnn_total_loss
parser.add_argument('--data_dir',
dest='data_dir',
help='data path',
default='/root/userfolder/yx/ch4_test_images',
type=str)
parser.add_argument('--res_dir',
dest='res_dir',
help='results path',
default='./txt_output',
type=str)
if len(sys.argv) == 1:
parser.print_help()
sys.exit(1)
args = parser.parse_args()
return args
if __name__ == '__main__':
args = parse_args()
print('Called with args:')
print(args)
det_net = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=False)
inference(det_net,
data_dir=args.data_dir,
res_dir=args.res_dir,
save_res=False)
def train():
faster_rcnn = build_whole_network.DetectionNetwork(
base_network_name=cfgs.NET_NAME, is_training=True)
with tf.name_scope('get_batch'):
img_name_batch, img_batch, gtboxes_and_label_batch, num_objects_batch = \
next_batch(dataset_name=cfgs.DATASET_NAME, # 'pascal', 'coco'
batch_size=cfgs.BATCH_SIZE,
shortside_len=cfgs.IMG_SHORT_SIDE_LEN,
is_training=True)
gtboxes_and_label = tf.reshape(gtboxes_and_label_batch, [-1, 5])
biases_regularizer = tf.no_regularizer
weights_regularizer = tf.contrib.layers.l2_regularizer(cfgs.WEIGHT_DECAY)
# list as many types of layers as possible, even if they are not used now
with slim.arg_scope([
slim.conv2d, slim.conv2d_in_plane, slim.conv2d_transpose,
slim.separable_conv2d, slim.fully_connected
],
weights_regularizer=weights_regularizer,
biases_regularizer=biases_regularizer,
biases_initializer=tf.constant_initializer(0.0)):
# result_dict, losses_dict = faster_rcnn.build_whole_detection_network(input_img_batch=img_batch,
# gtboxes_batch=gtboxes_and_label)
result_dict, losses_dict = faster_rcnn.build_whole_detection_network(
input_img_batch=img_batch, gtboxes_batch=gtboxes_and_label)
# ----------------------------------------------------------------------------------------------------build loss
weight_decay_loss = tf.add_n(slim.losses.get_regularization_losses())
# weight_decay_loss = tf.add_n(tf.losses.get_regularization_losses())
bbox_loss_m1 = losses_dict['bbox_loss_m1']
cls_loss_m1 = losses_dict['cls_loss_m1']
total_loss_m1 = bbox_loss_m1 + cls_loss_m1
bbox_loss_m2 = losses_dict['bbox_loss_m2']
cls_loss_m2 = losses_dict['cls_loss_m2']
total_loss_m2 = bbox_loss_m2 + cls_loss_m2
bbox_loss_m3 = losses_dict['bbox_loss_m3']
cls_loss_m3 = losses_dict['cls_loss_m3']
total_loss_m3 = bbox_loss_m3 + cls_loss_m3
total_loss = total_loss_m1 + total_loss_m2 + total_loss_m3 + weight_decay_loss
# ---------------------------------------------------------------------------------------------------add summary
tf.summary.scalar('SSH_M1_LOSS/cls_loss_m1', cls_loss_m1)
tf.summary.scalar('SSH_M1_LOSS/bbox_loss_m1', bbox_loss_m1)
tf.summary.scalar('SSH_M1_LOSS/total_loss_m1', total_loss_m1)
tf.summary.scalar('SSH_M2_LOSS/cls_loss_m2', cls_loss_m2)
tf.summary.scalar('SSH_M2_LOSS/bbox_loss_m2', bbox_loss_m2)
tf.summary.scalar('SSH_M2_LOSS/total_loss_m2', total_loss_m2)
tf.summary.scalar('SSH_M3_LOSS/cls_loss_m3', cls_loss_m3)
tf.summary.scalar('SSH_M3_LOSS/bbox_loss_m3', bbox_loss_m3)
tf.summary.scalar('SSH_M3_LOSS/total_loss_m3', total_loss_m3)
tf.summary.scalar('LOSS/total_loss', total_loss)
tf.summary.scalar('LOSS/regular_weights', weight_decay_loss)
gtboxes_in_img = show_box_in_tensor.draw_boxes_with_categories(
img_batch=img_batch,
boxes=gtboxes_and_label[:, :-1],
labels=gtboxes_and_label[:, -1])
if cfgs.ADD_BOX_IN_TENSORBOARD:
detections_in_img_m1 = \
show_box_in_tensor.draw_boxes_with_categories_and_scores(img_batch=img_batch,
boxes=result_dict['final_bbox_m1'],
labels=result_dict['final_category_m1'],
scores=result_dict['final_scores_m1'])
tf.summary.image('Compare/final_detection_m1', detections_in_img_m1)
detections_in_img_m2 = \
show_box_in_tensor.draw_boxes_with_categories_and_scores(img_batch=img_batch,
boxes=result_dict['final_bbox_m2'],
labels=result_dict['final_category_m2'],
scores=result_dict['final_scores_m2'])
tf.summary.image('Compare/final_detection_m2', detections_in_img_m2)
detections_in_img_m3 = \
show_box_in_tensor.draw_boxes_with_categories_and_scores(img_batch=img_batch,
boxes=result_dict['final_bbox_m3'],
labels=result_dict['final_category_m3'],
scores=result_dict['final_scores_m3'])
tf.summary.image('Compare/final_detection_m3', detections_in_img_m3)
tf.summary.image('Compare/gtboxes', gtboxes_in_img)
global_step = slim.get_or_create_global_step()
lr = tf.train.piecewise_constant(
global_step,
boundaries=[
np.int64(cfgs.DECAY_STEP[0]),
np.int64(cfgs.DECAY_STEP[1])
],
values=[cfgs.LR, cfgs.LR / 10., cfgs.LR / 100.])
tf.summary.scalar('lr', lr)
optimizer = tf.train.MomentumOptimizer(lr, momentum=cfgs.MOMENTUM)
# ---------------------------------------------------------------------------------------------compute gradients
gradients = faster_rcnn.get_gradients(optimizer, total_loss)
# enlarge_gradients for bias
if cfgs.MUTILPY_BIAS_GRADIENT:
gradients = faster_rcnn.enlarge_gradients_for_bias(gradients)
if cfgs.GRADIENT_CLIPPING_BY_NORM:
with tf.name_scope('clip_gradients'):
gradients = slim.learning.clip_gradient_norms(
gradients, cfgs.GRADIENT_CLIPPING_BY_NORM)
# train_op
train_op = optimizer.apply_gradients(grads_and_vars=gradients,
global_step=global_step)
summary_op = tf.summary.merge_all()
init_op = tf.group(tf.global_variables_initializer(),
tf.local_variables_initializer())
restorer, restore_ckpt = faster_rcnn.get_restorer()
saver = tf.train.Saver(max_to_keep=30)
config = tf.ConfigProto()
# config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(init_op)
if not restorer is None:
restorer.restore(sess, restore_ckpt)
print('restore model')
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess, coord)
summary_path = os.path.join(cfgs.SUMMARY_PATH, cfgs.VERSION)
if not os.path.exists(summary_path):
os.makedirs(summary_path)
summary_writer = tf.summary.FileWriter(summary_path, graph=sess.graph)
for step in range(cfgs.MAX_ITERATION):
training_time = time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time.time()))
# start = time.time()
# _, global_stepnp, img_name, totalLoss, summary_str = \
# sess.run(
# [train_op, global_step, img_name_batch, total_loss, summary_op])
#
# end = time.time()
#
# print(""" {}: step{} image_name:{} |\t total_loss:{} |\t per_cost_time:{}s""" \
# .format(training_time, global_stepnp, str(img_name[0]), totalLoss,
# (end - start)))
# summary_writer.add_summary(summary_str, global_stepnp)
# summary_writer.flush()
if step % cfgs.SHOW_TRAIN_INFO_INTE != 0 and step % cfgs.SMRY_ITER != 0:
_, global_stepnp = sess.run([train_op, global_step])
else:
if step % cfgs.SHOW_TRAIN_INFO_INTE == 0 and step % cfgs.SMRY_ITER != 0:
start = time.time()
_, global_stepnp, img_name, totalLoss = \
sess.run(
[train_op, global_step, img_name_batch, total_loss])
end = time.time()
print(""" {}: step{} image_name:{} |\t total_loss:{} |\t per_cost_time:{}s""" \
.format(training_time, global_stepnp, str(img_name[0]), totalLoss,
(end - start)))
else:
if step % cfgs.SMRY_ITER == 0:
_, global_stepnp, summary_str = sess.run(
[train_op, global_step, summary_op])
summary_writer.add_summary(summary_str, global_stepnp)
summary_writer.flush()
if (step > 0 and step % cfgs.SAVE_WEIGHTS_INTE
== 0) or (step == cfgs.MAX_ITERATION - 1):
save_dir = os.path.join(cfgs.TRAINED_CKPT, cfgs.VERSION)
if not os.path.exists(save_dir):
os.makedirs(save_dir)
save_ckpt = os.path.join(
save_dir, 'voc_' + str(global_stepnp) + 'model.ckpt')
saver.save(sess, save_ckpt)
print(' weights had been saved')
coord.request_stop()
coord.join(threads)