成功測試 tensorflow 2.10 於 windows11 使用 GPU

tensorflow>2.10 not support cuda

Caution: TensorFlow 2.10 was the last TensorFlow release that supported GPU on native-Windows. Starting with TensorFlow 2.11, you will need to install TensorFlow in WSL2, or install tensorflow or tensorflow-cpu and, optionally, try the **TensorFlow-DirectML-Plugin**
[https://www.tensorflow.org/install/pip#windows-native](https://www.tensorflow.org/install/pip#windows-native)
[Enable GPU acceleration for TensorFlow 2 with tensorflow-directml-plugin](https://learn.microsoft.com/en-us/windows/ai/directml/gpu-tensorflow-plugin)

conda create --name tfdml_plugin python=3.9 
conda activate tfdml_plugin

git clone https://github.com/microsoft/DirectML.git

[TensorFlow 2.0 DirectML Plugin on SqueezeNet](https://github.com/microsoft/DirectML/tree/master/TensorFlow/TF2/squeezenet)

conda create --name tfdml_plugin
conda activate tfdml_plugin
pip install tensorflow-cpu
pip install tensorflow-directml-plugin

python squeezenet.py --mode train --tb_profile --cifar10

執行訓練畫面
GPU使用統計
系統資源使用統計(Graphics_1)
args參數輸出(device='GPU:0')

[Build from source on Windows](https://www.tensorflow.org/install/source_windows)