TensorFlow导入失败及与Keras协同运行问题求助

问题详情 #

我的代码如下：

import os
os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'
import tensorflow as tf
from tensorflow.keras.layers import Conv1D, MaxPooling1D, GlobalAveragePooling1D, Dense, Dropout, BatchNormalization
from tensorflow.keras.models import Sequential
from tensorflow.keras.callbacks import EarlyStopping

在Visual Studio Code中运行时，TensorFlow相关模块无法正常导入，触发以下错误：

traceback (most recent call last):
  File "c:\Users\aliabzakh77\OneDrive\Desktop\PPG CNN\PPG Hypertension CNN.py", line 16, in <module>
    import tensorflow as tf
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\__init__.py", line 55, in <module>
    from tensorflow._api.v2 import compat
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\__init__.py", line 8, in <module>
    from tensorflow._api.v2.compat import v1
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\v1\__init__.py", line 30, in <module>
    from tensorflow._api.v2.compat.v1 import compat
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\v1\compat\__init__.py", line 8, in <module>    
    from tensorflow._api.v2.compat.v1.compat import v1
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\v1\compat\v1\__init__.py", line 47, in <module>
    from tensorflow._api.v2.compat.v1 import lite
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\v1\lite\__init__.py", line 9, in <module>      
    from tensorflow._api.v2.compat.v1.lite import experimental
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\v1\lite\experimental\__init__.py", line 8, in <module>
    from tensorflow._api.v2.compat.v1.lite.experimental import authoring
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\_api\v2\compat\v1\lite\experimental\authoring\__init__.py", line 8, in <module>
    from tensorflow.lite.python.authoring.authoring import compatible # line: 263
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\lite\python\authoring\authoring.py", line 42, in <module>     
    from tensorflow.lite.python import convert
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\lite\python\convert.py", line 27, in <module>
    from tensorflow.compiler.mlir.lite.python import wrap_converter
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\compiler\mlir\lite\python\wrap_converter.py", line 20, in <module>
    from tensorflow.compiler.mlir.quantization.tensorflow.python import py_function_lib
  File "C:\Users\aliabzakh77\AppData\Local\Programs\Python\Python311\Lib\site-packages\tensorflow\compiler\mlir\quantization\tensorflow\python\py_function_lib.py", line 29, in <module>
    from tensorflow.compiler.mlir.quantization.tensorflow.python import pywrap_function_lib

ImportError: DLL load failed while importing pywrap_function_lib: The specified procedure could not be found.

已尝试更新pip和TensorFlow，问题仍未解决，求可行解决方案。

解决方案 #

• 降级Python版本：TensorFlow 2.x对Python 3.11的兼容性存在不稳定情况，建议降级到Python 3.9或3.10，这两个版本是TensorFlow官方推荐的稳定适配版本。
• 彻底卸载重装TensorFlow：
  1. 卸载现有TensorFlow：pip uninstall tensorflow -y
  2. 清理pip缓存：pip cache purge
  3. 安装指定稳定版本（如2.15.0）：pip install tensorflow==2.15.0
• 安装Microsoft Visual C++ Redistributable：该错误多因系统缺失必要的C运行库导致，下载并安装对应系统位数的Microsoft Visual C Redistributable for Visual Studio 2015-2022。
• 修复环境变量：确保Python的Scripts目录和TensorFlow安装目录已添加到系统PATH中；可临时在代码开头添加以下语句测试：

  os.environ['PATH'] += ';' + os.path.dirname(tf.__file__) + '\\lite\\python\\authoring'
  

• 使用虚拟环境隔离依赖：
  1. 创建虚拟环境：python -m venv tf_env
  2. 激活虚拟环境：Windows执行tf_env\Scripts\activate，Linux/macOS执行source tf_env/bin/activate
  3. 在虚拟环境内安装TensorFlow：pip install tensorflow==2.15.0

内容的提问来源于stack exchange，提问作者Ali Abzakh