本地虚拟环境运行TensorFlow目标检测测试代码遇模块缺失错误

已安装所有所需依赖包，代码在Colaboratory中可正常运行，但在本地PC的虚拟环境中执行命令python object_detection/builders/model_builder_tf2_test.py时出现模块缺失错误，具体报错信息如下：

Traceback (most recent call last):
  File "C:\Users\cdelaney\Documents\TensorflowTraining\tfod\Lib\site-packages\object_detection\models\ssd_efficientnet_bifpn_feature_extractor.py", line 36, in <module>
    from official.legacy.image_classification.efficientnet import efficientnet_model
ModuleNotFoundError: No module named 'official.legacy'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "C:\Users\cdelaney\Documents\TensorflowTraining\models\research\object_detection\builders\model_builder_tf2_test.py", line 24, in <module>
    from object_detection.builders import model_builder
  File "C:\Users\cdelaney\Documents\TensorflowTraining\tfod\Lib\site-packages\object_detection\builders\model_builder.py", line 70, in <module>
    from object_detection.models import ssd_efficientnet_bifpn_feature_extractor as ssd_efficientnet_bifpn
  File "C:\Users\cdelaney\Documents\TensorflowTraining\tfod\Lib\site-packages\object_detection\models\ssd_efficientnet_bifpn_feature_extractor.py", line 38, in <module>
    from official.vision.image_classification.efficientnet import efficientnet_model
  File "C:\Users\cdelaney\Documents\TensorflowTraining\tfod\Lib\site-packages\official\vision\image_classification\efficientnet\efficientnet_model.py", line 37, in <module>
    from official.vision.image_classification import preprocessing
  File "C:\Users\cdelaney\Documents\TensorflowTraining\tfod\Lib\site-packages\official\vision\image_classification\preprocessing.py", line 25, in <module>
    from official.vision.image_classification import augment
  File "C:\Users\cdelaney\Documents\TensorflowTraining\tfod\Lib\site-packages\official\vision\image_classification\augment.py", line 31, in <module>
    from tensorflow.python.keras.layers.preprocessing import image_preprocessing as image_ops
ModuleNotFoundError: No module named 'tensorflow.python.keras.layers.preprocessing'

核心原因 #

问题根源是本地虚拟环境的TensorFlow版本与Object Detection API依赖的版本不兼容：

• tensorflow.python.keras.layers.preprocessing模块路径在不同TF版本中存在结构差异，本地版本过高或过低都会导致找不到该模块；
• official.legacy模块缺失说明TensorFlow官方Models库的版本和当前TF版本不匹配。

解决方案 #

1. 对齐Colab的TensorFlow版本 #

先在Colab中执行!pip show tensorflow查看当前使用的TF版本，然后在本地虚拟环境安装完全一致的版本：

pip install tensorflow==x.x.x  # 替换成Colab显示的版本号，例如2.15.0

2. 重新安装兼容的Object Detection依赖 #

卸载现有冲突包，再安装适配版本：

pip uninstall -y object-detection tensorflow-models-official
# 安装与TF版本匹配的官方Models包
pip install tensorflow-models-official==对应版本号

或者直接从TensorFlow Models仓库克隆对应分支的代码，本地编译安装：

git clone https://github.com/tensorflow/models.git
cd models/research
# 切换到与TF版本兼容的分支（如TF2.15对应r2.15分支）
git checkout r2.15
# 编译proto文件
protoc object_detection/protos/*.proto --python_out=.
# 安装Object Detection API
cp object_detection/packages/tf2/setup.py .
python -m pip install .

3. 验证修复 #

重新运行测试命令，确认错误消失：

python object_detection/builders/model_builder_tf2_test.py

内容的提问来源于stack exchange，提问作者Conor Delaney