chat-python/工程公司RAG部署指南.md

工程公司RAG部署指南

mineru部署：

conda create -n mineru python=3.10 conda activate mineru pip install uv uv pip install mineru[all]

设置国内路径：export MINERU_MODEL_SOURCE=modelscope 启动并自动下载模型：mineru-vllm-server --port 9999

如需使用其它端口，请修改./rag/document_load/pdf_load.py中server_url的默认值

RAG虚拟环境安装

conda create -n app python=3.11 conda activate app pip install -r requirements.txt

torch单独安装：pip install torch==2.8.0 torchvision==0.19.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cu121

保存下面内容为requirements.txt文件 accelerate==1.10.1 aiofiles==24.1.0 aiohappyeyeballs==2.6.1 aiohttp==3.11.18 aiosignal==1.3.2 aiosqlite==0.21.0 albucore==0.0.23 albumentations==2.0.5 annotated-types==0.7.0 antlr4-python3-runtime==4.9.3 anyio==4.9.0 argon2-cffi==23.1.0 argon2-cffi-bindings==21.2.0 astor==0.8.1 attrs==25.3.0 beautifulsoup4==4.14.2 blake3==1.0.8 boto3==1.38.1 botocore==1.38.1 Brotli==1.1.0 cachetools==6.2.1 cbor2==5.7.0 certifi==2025.1.31 cffi==1.17.1 chardet==5.2.0 charset-normalizer==3.4.1 click==8.2.0 cloudpickle==3.1.1 colorama==0.4.6 coloredlogs==15.0.1 colorlog==6.9.0 compressed-tensors==0.11.0 contourpy==1.3.2 Crawl4AI==0.6.3 cryptography==44.0.2 cssselect==1.3.0 cupy-cuda12x==13.6.0 cycler==0.12.1 dataclasses-json==0.6.7 depyf==0.19.0 dill==0.4.0 diskcache==5.6.3 distro==1.9.0 dnspython==2.8.0 doclayout_yolo==0.0.4 edge-tts==7.0.1 einops==0.8.1 email-validator==2.3.0 fake-http-header==0.3.5 fake-useragent==2.2.0 fast-langdetect==0.2.5 fastapi==0.115.12 fastapi-cli==0.0.13 fastapi-cloud-cli==0.3.1 fastrlock==0.8.3 fasttext-predict==0.9.2.4 ffmpy==0.6.3 filelock==3.18.0 flatbuffers==25.2.10 fonttools==4.57.0 frozendict==2.4.6 frozenlist==1.6.0 fsspec==2025.3.2 ftfy==6.3.1 gguf==0.17.1 gradio==5.49.1 gradio_client==1.13.3 gradio_pdf==0.0.22 greenlet==3.2.1 groovy==0.1.2 grpcio==1.67.1 h11==0.14.0 hf-xet==1.1.10 httpcore==1.0.8 httptools==0.7.1 httpx==0.28.1 httpx-sse==0.4.0 huggingface-hub==0.35.3 humanfriendly==10.0 humanize==4.12.3 idna==3.10 imageio==2.37.0 importlib_metadata==8.7.0 interegular==0.3.3 Jinja2==3.1.6 jiter==0.9.0 jmespath==1.0.1 joblib==1.4.2 json_repair==0.52.1 jsonpatch==1.33 jsonpointer==3.0.0 jsonschema==4.24.0 jsonschema-specifications==2025.4.1 kiwisolver==1.4.8 langchain==0.3.22 langchain-community==0.3.18 langchain-core==0.3.56 langchain-text-splitters==0.3.8 langsmith==0.3.34 lark==1.2.2 lazy_loader==0.4 litellm==1.72.0 llguidance==0.7.30 llvmlite==0.44.0 lm-format-enforcer==0.11.3 loguru==0.7.3 lxml==5.4.0 magic-pdf==1.3.10 magika==0.6.2 Markdown==3.8 markdown-it-py==3.0.0 MarkupSafe==3.0.2 marshmallow==3.26.1 matplotlib==3.10.1 mdurl==0.1.2 milvus-lite==2.4.12 milvus-model==0.2.12 milvus_cli==1.0.2 mineru==2.5.4 mineru_vl_utils==0.1.13 minio==7.2.15 mistral_common==1.8.5 modelscope==1.31.0 mpmath==1.3.0 msgpack==1.1.2 msgspec==0.19.0 multidict==6.4.3 mypy_extensions==1.1.0 mysql-connector-python==9.3.0 nest-asyncio==1.6.0 networkx==3.4.2 ninja==1.13.0 nltk==3.9.1 numba==0.61.2 numpy==1.26.4 nvidia-cublas-cu12==12.8.4.1 nvidia-cuda-cupti-cu12==12.8.90 nvidia-cuda-nvrtc-cu12==12.8.93 nvidia-cuda-runtime-cu12==12.8.90 nvidia-cudnn-cu12==9.10.2.21 nvidia-cufft-cu12==11.3.3.83 nvidia-cufile-cu12==1.13.1.3 nvidia-curand-cu12==10.3.9.90 nvidia-cusolver-cu12==11.7.3.90 nvidia-cusparse-cu12==12.5.8.93 nvidia-cusparselt-cu12==0.7.1 nvidia-nccl-cu12==2.27.3 nvidia-nvjitlink-cu12==12.8.93 nvidia-nvtx-cu12==12.8.90 ollama==0.4.8 omegaconf==2.3.0 onnxruntime==1.21.1 openai==1.109.1 openai-harmony==0.0.4 opencv-python==4.11.0.86 opencv-python-headless==4.11.0.86 orjson==3.10.16 outlines_core==0.2.11 packaging==24.2 pandas==2.2.3 partial-json-parser==0.2.1.1.post6 pdfminer.six==20250506 pdfplumber==0.11.6 pdftext==0.6.3 pillow==11.3.0 playwright==1.52.0 prometheus-fastapi-instrumentator==7.1.0 prometheus_client==0.23.1 propcache==0.3.1 protobuf==6.30.2 psutil==7.0.0 py-cpuinfo==9.0.0 pybase64==1.4.2 pyclipper==1.3.0.post6 pycountry==24.6.1 pycparser==2.22 pycryptodome==3.22.0 pydantic==2.11.10 pydantic-extra-types==2.10.6 pydantic-settings==2.9.1 pydantic_core==2.33.2 pydub==0.25.1 pyee==13.0.0 Pygments==2.19.1 pymilvus==2.5.4 PyMuPDF==1.24.14 pyOpenSSL==25.1.0 pyparsing==3.2.3 pypdf==6.1.1 pypdfium2==4.30.0 pyperclip==1.9.0 python-dateutil==2.9.0.post0 python-docx==1.1.2 python-dotenv==1.1.0 python-json-logger==4.0.0 python-multipart==0.0.20 pytz==2025.2 PyYAML==6.0.2 pyzmq==27.1.0 rank-bm25==0.2.2 rapid-table==1.0.5 ray==2.50.1 redis==5.2.1 referencing==0.36.2 regex==2024.11.6 reportlab==4.4.4 requests==2.32.3 requests-toolbelt==1.0.0 rich==14.0.0 rich-toolkit==0.15.1 rignore==0.7.1 robust-downloader==0.0.2 rpds-py==0.25.1 ruff==0.14.1 s3transfer==0.12.0 safehttpx==0.1.6 safetensors==0.5.3 scikit-image==0.25.2 scikit-learn==1.6.1 scipy==1.15.2 seaborn==0.13.2 semantic-version==2.10.0 sentence-transformers==4.1.0 sentencepiece==0.2.1 sentry-sdk==2.42.0 setproctitle==1.3.7 shapely==2.1.0 shellingham==1.5.4 simsimd==6.2.1 six==1.17.0 sniffio==1.3.1 snowballstemmer==2.2.0 soundfile==0.13.1 soupsieve==2.7 soxr==1.0.0 SQLAlchemy==2.0.40 srt==3.5.3 sse-starlette==2.3.3 starlette==0.46.2 stringzilla==3.12.5 sympy==1.13.3 tabulate==0.8.9 tenacity==9.1.2 tf-playwright-stealth==1.1.2 thop==0.1.1.post2209072238 threadpoolctl==3.6.0 tifffile==2025.10.16 tiktoken==0.9.0 tokenizers==0.22.1 tomlkit==0.13.3 tqdm==4.67.1 transformers==4.57.1 triton==3.4.0 typer==0.19.2 typing-inspect==0.9.0 typing-inspection==0.4.0 typing_extensions==4.13.2 tzdata==2025.2 ujson==5.10.0 ultralytics==8.3.115 ultralytics-thop==2.0.14 urllib3==2.4.0 uv==0.9.3 uvicorn==0.34.2 uvloop==0.22.1 vllm==0.10.2 watchfiles==1.1.1 wcwidth==0.2.13 websockets==15.0.1 xformers==0.0.32.post1 xgrammar==0.1.23 xxhash==3.5.0 yarl==1.20.0 zipp==3.22.0 zstandard==0.23.0

其它：

嵌入模型：https://www.modelscope.cn/models/intfloat/multilingual-e5-large-instruct 重排序模型：https://www.modelscope.cn/models/maidalun/bce-reranker-base_v1 下载后修改目录下的./rag/load_model.py中的模型路径

修改./config中的相应配置

因为运行环境的不同指定了不同的GPU，可在./rag_server.py顶部指定os.environ["CUDA_VISIBLE_DEVICES"] 以及./rag/load_model.py中的device = "cuda:1" if torch.cuda.is_available() else "cpu"的cuda:1 为 cuda:x

运行rag_server.py 默认6666端口