Windows11下Python3.12安装Keras及LSTM RNN支持失败求助

问题描述 #

Windows11华硕笔记本，VSCode环境中为Python3.12安装Keras时，出现dm-tree编译失败的报错，核心信息如下：

File "C:\Users\João\AppData\Local\Programs\Python\Python312\Lib\subprocess.py", line 413, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '["cmake", "C:\\Users\\João\\AppData\\Local\\Temp\\pip-install-ll8f_kl9\\dm-tree_85d7473578b94b2bb03305ca6775c834\\tree", "-DPython3_ROOT_DIR=C:\\Users\\João\\AppData\\Local\\Programs\\Python\\Python312", "-DPython3_EXECUTABLE=C:\\Users\\João\\AppData\\Local\\Programs\\Python\\Python312\\python.exe", "-DCMAKE_LIBRARY_OUTPUT_DIRECTORY=C:\\Users\\João\\AppData\\Local\\Temp\\pip-install-ll8f_kl9\\dm-tree_85d7473578b94b2bb03305ca6775c834\\build\\lib.win-amd64-cpython-312", "-DCMAKE_BUILD_TYPE=Release"]' returned non-zero exit status 1.
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for dm-tree
Running setup.py clean for dm-tree
Failed to build dm-tree
ERROR: Could not build wheels for dm-tree, which is required to install pyproject.toml-based projects

已尝试降级Python、安装TensorFlow、CMD命令行安装、配置NVIDIA CUDA，但问题未解决，需尽快完成Keras安装用于大学课程展示。

解决方案 #

1. 直接安装预编译的dm-tree wheel包（优先推荐） #

跳过编译环节，直接使用适配版本的预编译wheel包：

• 找到对应Python3.12+Windows amd64架构的dm-tree wheel文件
• 执行以下命令安装（替换为你的文件实际路径）：

  pip install C:\下载路径\dm_tree-0.1.8-cp312-cp312-win_amd64.whl
  

• 完成后执行pip install keras即可完成安装。

2. 修复CMake环境配置 #

如果需要自行编译，需确保CMake环境正确：

• 从CMake官网下载Windows版本安装包，安装时勾选Add CMake to the system PATH for all users
• 重启VSCode或终端，执行cmake --version验证环境生效
• 重新执行pip install keras

3. 彻底切换到Python3.11环境（备选） #

Python3.12部分依赖的编译支持尚未完善，可尝试：

• 卸载Python3.12，安装Python3.11（安装时务必勾选Add Python to PATH）
• 创建并激活虚拟环境：

  python -m venv keras_env
  keras_env\Scripts\activate
  

• 安装TensorFlow（自带Keras组件）：

  pip install tensorflow
  

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