ValueError: Layer #222

Question

环境：
tensorflow 2.4.3
keras 2.4.0
python 3.8.3
numpy 1.19.2
skimage 0.19

代码：

import os
import matplotlib.pyplot as plt
from mrcnn.config import Config
from mrcnn import model as modellib, utils
from mrcnn.visualize import display_instances


# 设置数据路径
data_dir = 'fruits-360'
train_dir = os.path.join(data_dir, 'Training')
validation_dir = os.path.join(data_dir, 'Validation')


# 配置Mask R-CNN模型
class MyConfig(Config):
    NAME = "my_config"
    GPU_COUNT = 1
    IMAGES_PER_GPU = 1
    NUM_CLASSES = 6
    IMAGE_MIN_DIM = 640
    IMAGE_MAX_DIM = 640


config = MyConfig()

# 创建Mask R-CNN模型
model = modellib.MaskRCNN(mode="inference", config=config, model_dir=os.getcwd())

# 加载预训练权重文件
weights_path = "mask_rcnn_balloon.h5"
model.load_weights(weights_path, by_name=True)

# 准备要识别的图片
image_paths = ['fruits-360/Test/Apple Golden 1/63_100.jpg', 'fruits-360/Test/Apple Golden 1/72_100.jpg']  # 替换为您要识别的图片路径

class_names = ['Apple Golden 1', 'Apple Golden 2', 'Apple Golden 3', 'Apple Red 1', 'Apple Red 2', 'Apple Red 3']

# 进行目标识别
for image_path in image_paths:
    image = plt.imread(image_path)
    results = model.detect([image], verbose=1)

    # 可视化识别结果
    r = results[0]
    display_instances(image, r['rois'], r['masks'], r['class_ids'], class_names, r['scores'])
    plt.show()

在使用mrcnn模型时出现了许多问题，我也在stackflow上找到了一个解决方案，但按解决方案的步骤去做后，仍然会出现问题。
查询的解决方案如下：

Installing tensorflow with version as following

pip uninstall tensorflow -y
pip uninstall keras -y
pip install tensorflow==2.4.3
pip install keras==2.4.0
After above, some errors will arise. You could solve them by following steps.

@Error: [module 'tensorflow' has no attribute XXXXXXXX]

In the model.py or your code, resolving some api with tf.compat.v1, e.g. tf.compat.v1.Session or import tensorflow.compat.v1 as tf

@Error: [ValueError: Tried to convert 'shape' to a tensor and failed. Error: None values not supported.]

mrcnn_bbox = KL.Reshape((-1, num_classes, 4), name="mrcnn_bbox")(x)
replace with this this if-else code block:

if s[1]==None:
    mrcnn_bbox = KL.Reshape((-1, num_classes, 4), name="mrcnn_bbox")(x)
else:
    mrcnn_bbox = KL.Reshape((s[1], num_classes, 4), name="mrcnn_bbox")(x)
@Error: [ValueError: None values not supported.]

indices = tf.stack([tf.range(probs.shape[0]), class_ids], axis=1)
replace with

indices = tf.stack([tf.range(tf.shape(probs)[0]), class_ids], axis = 1)
@Error: [AttributeError: module 'keras.engine.saving' has no attribute 'load_weights_from_hdf5_group_by_name']

from keras import saving
replace with

from tensorflow.python.keras.saving import hdf5_format
and

saving.load_weights_from_hdf5_group(f, layers)
saving.load_weights_from_hdf5_group_by_name(f, layers)
replace with

hdf5_format.load_weights_from_hdf5_group(f, layers)
hdf5_format.load_weights_from_hdf5_group_by_name(f, layers)

做完最后一步之后，出现问题：

File "D:/HomeWork/deep_learning/yolox.py", line 31, in
model.load_weights(weights_path, by_name=True)
File "C:\Program Files\Python38\lib\site-packages\mrcnn\model.py", line 2122, in load_weights
hdf5_format.load_weights_from_hdf5_group(f, layers)
File "C:\Users\86138\AppData\Roaming\Python\Python38\site-packages\tensorflow\python\keras\saving\hdf5_format.py", line 702, in load_weights_from_hdf5_group
raise ValueError('Layer #' + str(k) + ' (named "' + layer.name +
ValueError: Layer #222 (named "mrcnn_bbox_fc" in the current model) was found to correspond to layer mrcnn_class_bn1 in the save file. However the new layer mrcnn_bbox_fc expects 2 weights, but the saved weights have 4 elements.

Answer 1

引自强大而免费的微信小程序：皆我百晓生。

---

这个错误是由于保存的模型权重与加载的模型结构不匹配导致的。具体来说，模型中的第222个层被命名为"mrcnn_bbox_fc"，但是保存的权重中对应的层被命名为"mrcnn_class_bn1"。

要解决这个问题，你可以尝试以下方法之一：

1. 使用相同的模型结构重新训练并保存权重：重新训练模型，并确保在保存模型权重时使用相同的命名方式。然后使用重新训练的权重进行预测。

2. 修改加载权重的代码：你可以在加载权重之前检查模型中每个层的名称和顺序，然后根据需要重命名加载的权重。例如，你可以使用model.summary()查看模型结构，并在加载权重之前通过修改代码中的层名称来对齐。

也可以尝试更新mrcnn库的版本来解决该问题。请尝试更新mrcnn库的版本，或查看是否有其他版本兼容的解决方案可用。

希望这些信息对你有帮助，如果还有其他问题，请随时询问。