ValueError: Error when checking input: expected input_image_meta to have shape (None, 14) but got ar-优快云博客

本文链接：https://blog.youkuaiyun.com/oMoDao1/article/details/83860101

本文详细解析了在使用Mask R-CNN进行训练时遇到的常见错误：输入元数据形状不匹配。深入探讨了错误原因及解决方案，指出在增加新类别时需相应调整NUM_CLASSES参数。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

在跑maskrcnn的时候出现以下错误：

Traceback (most recent call last):
File "/home/ubuntu/DNN_Projection/Mask_RCNN--Training/samples/nucleus/nucleus.py", line 483, in <module>
    train(model, args.dataset, args.subset)
File "/home/ubuntu/DNN_Projection/Mask_RCNN--Training/samples/nucleus/nucleus.py", line 285, in train
    layers='heads')
File "/home/ubuntu/DNN_Projection/Mask_RCNN--Training/mrcnn/model.py", line 2375, in train
    use_multiprocessing=True,
File "/home/ubuntu/anaconda3/envs/tensorflow1.3/lib/python3.6/site-packages/keras/legacy/interfaces.py", line 87, in wrapper
    return func(*args, **kwargs)
File "/home/ubuntu/anaconda3/envs/tensorflow1.3/lib/python3.6/site-packages/keras/engine/training.py", line 2042, in fit_generator
    class_weight=class_weight)
File "/home/ubuntu/anaconda3/envs/tensorflow1.3/lib/python3.6/site-packages/keras/engine/training.py", line 1756, in train_on_batch
    check_batch_axis=True)
File "/home/ubuntu/anaconda3/envs/tensorflow1.3/lib/python3.6/site-packages/keras/engine/training.py", line 1378, in _standardize_user_data
    exception_prefix='input')
File "/home/ubuntu/anaconda3/envs/tensorflow1.3/lib/python3.6/site-packages/keras/engine/training.py", line 144, in _standardize_input_data
    str(array.shape))
ValueError: Error when checking input: expected input_image_meta to have shape (None, 14) but got array with shape (1, 15)

出现该错误是因为我新增了一类，例：

        self.add_class("nucleus", 1, "RBC")
        self.add_class("nucleus", 2, "ELSE2")

解决办法是修改NUM_CLASSES参数，增加了多少类，就对应的修改对应数值

NUM_CLASSES = 1 + 1