简述企业网站推广的一般策略,今天最新体育新闻,wordpress 4.8 主题,icp备案网站更名1. 登录PyTorch查看需要的Python和CUDA版本#xff0c;并进行安装 2.安装CUDA
下载并安装 CUDA Toolkit - Free Tools and Training | NVIDIA Developer
安装后运行 nvidia-smi #xff0c; 返回显卡各种信息 3.使用conda创建虚拟环境#xff08;安装过程中有些包会冲突…1. 登录PyTorch查看需要的Python和CUDA版本并进行安装 2.安装CUDA
下载并安装 CUDA Toolkit - Free Tools and Training | NVIDIA Developer
安装后运行 nvidia-smi 返回显卡各种信息 3.使用conda创建虚拟环境安装过程中有些包会冲突这是我倒出的关键包
# conda create -f environment.ymlname: aipy11
channels:- http://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main- http://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/win-64- http://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/win-64
dependencies:- bzip21.0.8h2bbff1b_6- ca-certificates2025.2.25haa95532_0- expat2.7.1h8ddb27b_0- libffi3.4.4hd77b12b_1- openssl3.0.16h3f729d1_0- pip25.1pyhc872135_2- python3.11.13h981015d_0- setuptools78.1.1py311haa95532_0- sqlite3.45.3h2bbff1b_0- tk8.6.14h5e9d12e_1- vc14.42haa95532_5- vs2015_runtime14.42.34433hbfb602d_5- wheel0.45.1py311haa95532_0- xz5.6.4h4754444_1- zlib1.2.13h8cc25b3_1- pip:- accelerate1.7.0- addict2.4.0- huggingface-hub0.33.0- langchain0.3.25- langchain-community0.3.25- langchain-core0.3.65- langchain-openai0.3.22- langchain-text-splitters0.3.8- langsmith0.3.45- llama-cloud0.1.26- llama-cloud-services0.6.31- llama-index0.12.42- llama-index-agent-openai0.4.11- llama-index-cli0.4.3- llama-index-core0.12.42- llama-index-embeddings-huggingface0.5.4- llama-index-embeddings-openai0.3.1- llama-index-indices-managed-llama-cloud0.7.7- llama-index-llms-huggingface0.5.0- llama-index-llms-openai0.4.5- llama-index-multi-modal-llms-openai0.5.1- llama-index-program-openai0.3.2- llama-index-question-gen-openai0.3.1- llama-index-readers-file0.4.9- llama-index-readers-llama-parse0.4.0- llama-parse0.6.31- pillow11.0.0- pydantic2.11.5- pydantic-core2.33.2- pydantic-settings2.9.1
3.大模型的下载
可以从modelscope、huggingface、ollama等平台下载 指定保存位置否则都到了c盘
os.environ[TRANSFORMERS_CACHE] cache_apath
os.environ[HF_HOME] cache_apath
os.environ[MODELSCOPE_CACHE] cache_apath
