Windows 安裝 TensorFlow GPU 版的問題

昨天在新的電競桌機安裝好 NVIDIA 顯示卡的 CUDA 函式庫後, 順利安裝好 GPU 版的 PyTorch. 想說接著就來安裝 GPU 版的 TensorFlow 吧! 指令如下 : 

pip install tensorflow-gpu

參考 :

# 使用 pip 安裝 TensorFlow 

但這回卻出現錯誤 :

C:\Users\USER>pip install tensorflow-gpu   

Collecting tensorflow-gpu

  Downloading tensorflow-gpu-2.12.0.tar.gz (2.6 kB)

  Preparing metadata (setup.py) ... done

Collecting python_version>"3.7" (from tensorflow-gpu)

  Downloading python_version-0.0.2-py2.py3-none-any.whl.metadata (1.7 kB)

Downloading python_version-0.0.2-py2.py3-none-any.whl (3.4 kB)

Building wheels for collected packages: tensorflow-gpu

  Building wheel for tensorflow-gpu (setup.py) ... error　　

  error: subprocess-exited-with-error

  × python setup.py bdist_wheel did not run successfully.

  │ exit code: 1

  ╰─> [18 lines of output]

      Traceback (most recent call last):

        File "<string>", line 2, in <module>

        File "<pip-setuptools-caller>", line 34, in <module>

        File "C:\Users\USER\AppData\Local\Temp\pip-install-k8fazmsx\tensorflow-gpu_379ac048cb9c49a387dd64f61929c840\setup.py", line 37, in <module>

          raise Exception(TF_REMOVAL_WARNING)

      Exception:

      =========================================================

      The "tensorflow-gpu" package has been removed!

      Please install "tensorflow" instead.

      Other than the name, the two packages have been identical

      since TensorFlow 2.1, or roughly since Sep 2019. For more

      information, see: pypi.org/project/tensorflow-gpu

      =========================================================

      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.

  ERROR: Failed building wheel for tensorflow-gpu

  Running setup.py clean for tensorflow-gpu

Failed to build tensorflow-gpu

ERROR: Could not build wheels for tensorflow-gpu, which is required to install pyproject.toml-based projects

搜尋 "tensorflow gpu subprocess-exited-with-error" 發現, 原來 tensorflow-gpu 套件已經不存在了, 安裝 tensorflow 套件即可, 因為 tensorflow 已支援 CUDA 加速運算, 參考 :

# https://pypi.org/project/tensorflow-gpu/