Mask R-CNN训练自己得数据集出错，布尔维度不匹配（已解决）

错误提示如下：

class_ids = class_ids[_idx]
IndexError: boolean index did not match indexed array along dimension 0; dimension is 0 but corresponding boolean dimension is 1

ERROR:root:Error processing image {'id': 6, 'source': 'shapes', 'path': 'G:\\Download\\MASKRCNN\\Mask_RCNN-master\\train_data\\pic/008.jpg', 'width': 480, 'height': 320, 'mask_path': 'G:\\Download\\MASKRCNN\\Mask_RCNN-master\\train_data\\cv2_mask/008.png', 'yaml_path': 'G:\\Download\\MASKRCNN\\Mask_RCNN-master\\train_data\\labelme_json/008_json/info.yaml'}
Traceback (most recent call last):

训练5类 ；add_class 0-4 ； NUM_CLASSES = 1 + 5；

NAME = "shapes"
    GPU_COUNT = 1
    IMAGES_PER_GPU = 2
    NUM_CLASSES = 1 + 5
    IMAGE_MIN_DIM = 256
    IMAGE_MAX_DIM = 256
    RPN_ANCHOR_SCALES = (8 * 6, 16 * 6, 32 * 6, 64 * 6, 128 * 6)
    TRAIN_ROIS_PER_IMAGE = 100
    STEPS_PER_EPOCH = 100
    VALIDATION_STEPS = 50
load shape部分
    self.add_class("shapes", 0, "D00")
    self.add_class("shapes", 1, "D10")
    self.add_class("shapes", 2, "D20")
    self.add_class("shapes", 3, "D30")
    self.add_class("shapes", 4, "D40")

mask大小为480*320，8位深度

有没有大神帮忙指点一下问题出在哪里~

————————————————分割——————————

原因在后面写标签部分，只改了前面，后面检测出来后标签没有对应上。

已解决