def gen(self, au=True, crop_size=64, iteration=None):
while True:
data_yield = [
self.index % self.n, (self.index + self.step) % self.n if
(self.index + self.step) < self.n else self.n
]
self.index = (self.index + self.step) % self.n
list_images_base = self.list_images_all[
data_yield[0]:data_yield[1]]
list_gt_base = self.list_gt_all[data_yield[0]:data_yield[1]]
list_images_base = pre_processing.my_PreProc(
list_images_base) # 图片增强预处理
list_images_aug = []
list_gt_aug = []
for image, gt in zip(list_images_base, list_gt_base):
if au:
if crop_size == prepare_dataset.DESIRED_DATA_SHAPE[0]:
for _ in range(self.repeat):
image, gt = data_augmentation.random_augmentation(
image, gt)
list_images_aug.append(image)
list_gt_aug.append(gt)
else:
image, gt = data_augmentation.random_augmentation(
image, gt)
list_images_aug.append(image)
list_gt_aug.append(gt)
else:
list_images_aug.append(image)
list_gt_aug.append(gt)
list_images = []
list_gt = []
if crop_size == prepare_dataset.DESIRED_DATA_SHAPE[0]:
list_images = list_images_aug
list_gt = list_gt_aug
else:
for image, gt in zip(list_images_aug, list_gt_aug):
for _ in range(self.repeat):
image_, gt_ = random_crop(image, gt, crop_size)
list_images.append(image_)
list_gt.append(gt_)
outs = {}
for iteration_id in range(iteration):
outs.update({f'out1{iteration_id + 1}': np.array(list_gt)})
outs.update({'final_out': np.array(list_gt)})
yield np.array(list_images), outs
def gen(self, au=True, crop_size=48, iteration=None):
while True:
data_yield = [self.index % self.n,
(self.index + self.step) % self.n if (self.index + self.step) < self.n else self.n]
self.index = (self.index + self.step) % self.n
list_images_base = self.list_images_all[data_yield[0]:data_yield[1]]
list_gt_base = self.list_gt_all[data_yield[0]:data_yield[1]]
list_gt_onehot_base = self.list_gt_all_onehot[data_yield[0]:data_yield[1]]
list_images_aug = []
list_gt_aug = []
list_gt_onehot_aug = []
image_id = -1
for image, gt in zip(list_images_base, list_gt_base):
image_id += 1
gt2 = list_gt_onehot_base[image_id]
if au:
if crop_size == prepare_dataset.DESIRED_DATA_SHAPE[0]:
for _ in range(self.repeat):
image, gt, gt2 = data_augmentation.random_augmentation(image, gt, gt2)
list_images_aug.append(image)
list_gt_aug.append(gt)
list_gt_onehot_aug.append(gt2)
else:
image, gt, gt2 = data_augmentation.random_augmentation(image, gt, gt2)
list_images_aug.append(image)
list_gt_aug.append(gt)
list_gt_onehot_aug.append(gt2)
else:
list_images_aug.append(image)
list_gt_aug.append(gt)
list_gt_onehot_aug.append(gt2)
list_images = []
list_gt = []
list_gt_onehot = []
image_id = -1
if crop_size == prepare_dataset.DESIRED_DATA_SHAPE[0]:
list_images = list_images_aug
list_gt = list_gt_aug
list_gt_onehot = list_gt_onehot_aug
else:
for image, gt in zip(list_images_aug, list_gt_aug):
image_id += 1
for _ in range(self.repeat):
image_, gt_, gt_onehot_ = random_crop(image, gt, list_gt_onehot_aug[image_id], crop_size)
list_images.append(image_)
list_gt.append(gt_)
list_gt_onehot.append(gt_onehot_)
outs = {}
for iteration_id in range(iteration):
outs.update({f'out1{iteration_id + 1}': np.array(list_gt)[:,0]})
outs.update({'seg_final_out': np.array(list_gt)[:,0]})
# outs.update({'crossing_final_out': np.array(list_gt)[:,1]})
outs.update({'cls_art_final_out': np.array(list_gt)[:,2]})
outs.update({'cls_vei_final_out': np.array(list_gt)[:,3]})
for iteration_id in range(iteration):
outs.update({f'out1_cls_art{iteration_id + 1}': np.array(list_gt)[:,2]})
for iteration_id in range(iteration):
outs.update({f'out1_cls_vei{iteration_id + 1}': np.array(list_gt)[:,3]})
yield np.array(list_images), outs