conda基础：GPU-torch和pytorch共存

安装langchain流程：

requirements.txt文件

内容：

torch==2.1.2
torchvision==0.16.2
torchaudio==2.1.2
xformers==0.0.23.post1
transformers==4.37.2
sentence_transformers==2.2.2
langchain==0.0.354
langchain-community==0.0.19
langchain-experimental==0.0.47
pydantic==1.10.13
fschat==0.2.35
openai==1.9.0
fastapi==0.109.0
sse_starlette==1.8.2
nltk==3.8.1
uvicorn>=0.27.0.post1
starlette==0.35.0
unstructured[all-docs] # ==0.11.8
python-magic-bin; sys_platform == 'win32'
SQLAlchemy==2.0.25
faiss-cpu==1.7.4
accelerate==0.24.1
spacy==3.7.2
PyMuPDF==1.23.16
rapidocr_onnxruntime==1.3.8
requests==2.31.0
pathlib==1.0.1
pytest==7.4.3
numexpr==2.8.6
strsimpy==0.2.1
markdownify==0.11.6
tiktoken==0.5.2
tqdm==4.66.1
websockets==12.0
numpy==1.24.4
pandas==2.0.3
einops==0.7.0
transformers_stream_generator==0.0.4
vllm==0.2.7; sys_platform == "linux"
llama-index==0.9.35

# jq==1.6.0
# beautifulsoup4==4.12.2
# pysrt==1.1.2
# dashscope==1.13.6 # qwen
# volcengine==1.0.134 # fangzhou
# uncomment libs if you want to use corresponding vector store
# pymilvus==2.3.6
# psycopg2==2.9.9
# pgvector>=0.2.4
# chromadb==0.4.13

#flash-attn==2.4.2 # For Orion-14B-Chat and Qwen-14B-Chat
#autoawq==0.1.8 # For Int4
#rapidocr_paddle[gpu]==1.3.11 # gpu accelleration for ocr of pdf and image files

arxiv==2.1.0
youtube-search==2.1.2
duckduckgo-search==3.9.9
metaphor-python==0.1.23

streamlit==1.30.0
streamlit-option-menu==0.3.12
streamlit-antd-components==0.3.1
streamlit-chatbox==1.1.11
streamlit-modal==0.1.0
streamlit-aggrid==0.3.4.post3

httpx==0.26.0
httpx_sse==0.4.0
watchdog==3.0.0
pyjwt==2.8.0
 

pip install -r .\requirements.txt

pip install torch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1或者

conda install pytorch==1.12.1 torchvision==0.13.1 torchaudio==0.12.1 cudatoolkit=11.3 -c pytorch

conda list查看安装包，存在torch2.1.2和与cuda11.3对应的pytorch1.12.1即可

不对应则修改requirement.txt文件