mask-r-cnn在得到mPA时遇到的情况

大神们好，这是我想得到模型的mAP时已经到了训练阶段遇到的问题，困惑了我好久，找不到解决的办法，如果有会的大神请赐教，有酬金，谢谢啦

KeyError                                  Traceback (most recent call last)
<ipython-input-85-83fb3ae74319> in <module>
      6             learning_rate=config.LEARNING_RATE,
      7             epochs=1,
----> 8             layers='heads')

D:\pythonprojects\lib\site-packages\mrcnn\model.py in train(self, train_dataset, val_dataset, learning_rate, epochs, layers, augmentation)
   2360             max_queue_size=100,
   2361             workers=workers,
-> 2362             use_multiprocessing=True,
   2363         )
   2364         self.epoch = max(self.epoch, epochs)

D:\pythonprojects\lib\site-packages\keras\legacy\interfaces.py in wrapper(*args, **kwargs)
     89                 warnings.warn('Update your `' + object_name + '` call to the ' +
     90                               'Keras 2 API: ' + signature, stacklevel=2)
---> 91             return func(*args, **kwargs)
     92         wrapper._original_function = func
     93         return wrapper

D:\pythonprojects\lib\site-packages\keras\engine\training.py in fit_generator(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, validation_freq, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)
   1656             use_multiprocessing=use_multiprocessing,
   1657             shuffle=shuffle,
-> 1658             initial_epoch=initial_epoch)
   1659 
   1660     @interfaces.legacy_generator_methods_support

D:\pythonprojects\lib\site-packages\keras\engine\training_generator.py in fit_generator(model, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, validation_freq, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)
    179             batch_index = 0
    180             while steps_done < steps_per_epoch:
--> 181                 generator_output = next(output_generator)
    182 
    183                 if not hasattr(generator_output, '__len__'):

D:\pythonprojects\lib\site-packages\mrcnn\model.py in data_generator(dataset, config, shuffle, augment, augmentation, random_rois, batch_size, detection_targets)
   1712                 load_image_gt(dataset, config, image_id, augment=augment,
   1713                               augmentation=augmentation,
-> 1714                               use_mini_mask=config.USE_MINI_MASK)
   1715 
   1716             # Skip images that have no instances. This can happen in cases

D:\pythonprojects\lib\site-packages\mrcnn\model.py in load_image_gt(dataset, config, image_id, augment, augmentation, use_mini_mask)
   1290     # classes supported in the dataset of this image.
   1291     active_class_ids = np.zeros([dataset.num_classes], dtype=np.int32)
-> 1292     source_class_ids = dataset.source_class_ids[dataset.image_info[image_id]["source"]]
   1293     active_class_ids[source_class_ids] = 1
   1294 

KeyError: 'shapes'
```python

困惑了好久有我在不用怕，你把报错复制出来发给我

参考一下

keyerror一般指的是你在使用字典里不存在的key产生的错误。
解题思路：
通过用Shape.getLocation方法，去给其中key的值，自行确定赋值
参考链接1：https://vimsky.com/examples/detail/python-ex-shape-Shape-getLocation-method.html
参考链接2：https://javajgs.com/archives/78572

mask rcnn测试中遇到的问题解决
https://blog.csdn.net/qq_34713831/article/details/85797622