# 安裝Tesla V100 for win server 2019 操作手冊


win srv 2019 tesla v100 driver download
https://www.nvidia.com/Download/Find.aspx
安裝
因為解壓縮根本沒反應，所以手動7-zip執行
安裝 anaconda 虛擬環境管理工具
https://www.anaconda.com/products/distribution

# 安裝 GPU cuda 11.6
(先)安裝 visual studio comnumication
https://developer.nvidia.com/cuda-11-6-0-download-archive?target_os=Windows&target_arch=x86_64&target_version=Server2019&target_type=exe_network
# 安裝 cudnn for cuda 11.6
登入 nvidia 下載 cudnn xxx .zip
解壓縮後放到 C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6 路徑下
用 nvidia-smi 得到下圖 成功

# 安裝 python

python -m pip install --upgrade pip 
conda update -n base -c defaults conda
conda env list  # 列出python 虛擬環境清單
conda create --name py3.10 python=3.10  # 建立新python環境

Preparing transaction: done
Verifying transaction: done
Executing transaction: done
#
# To activate this environment, use
#
#     $ conda activate py3.10
#
# To deactivate an active environment, use
#
#     $ conda deactivate
conda activate py3.10
conda install -c anaconda git
cd c:\develop
git clone https://github.com/camenduru/stable-diffusion-webui-colab
pip install jupyter
jupyter notebook

成功開啟 jupyter notebook

# 到目前為止都健康的，但是這個github裡面的程式都是for google colab的 .ipython 所以還是要在 google colab裡面執行

另外弄了一個，測試沒問題https://github.com/Belval/TextRecognitionDataGenerator
cd c:\develop\TextRecognitionDataGenerator\trdg
python run.py -c 10 -l cn -w 5 -sw 0 -or 1