SenseVoice
SenseVoice copied to clipboard
使用Dockerfile构建的镜像无法正常访问
Notice: In order to resolve issues more efficiently, please raise issue following the template. (注意:为了更加高效率解决您遇到的问题,请按照模板提问,补充细节)
🐛 Bug
To Reproduce
FROM pytorch/pytorch:2.3.0-cuda12.1-cudnn8-runtime
ENV DEBIAN_FRONTEND=noninteractive
EXPOSE 50000
ENV TZ=Asia/Shanghai
RUN mkdir -p /app
WORKDIR /app
RUN apt update && apt install -y wget curl net-tools tree git git-lfs
RUN git clone https://github.com/FunAudioLLM/SenseVoice.git
WORKDIR /app/SenseVoice
RUN pip install -r requirements.txt
RUN rm -rf /var/lib/apt/lists/*
CMD ["python", "webui.py"]
services:
sensevoice:
image: registry.cn-hangzhou.aliyuncs.com/joybo/sensevoice
container_name: sensevoice
hostname: sensevoice
environment:
- TZ=Asia/Shanghai
- HF_HOME=/app/sensevoice/models
- MODELSCOPE_CACHE=/app/sensevoice/models
- GRADIO_SERVER_NAME=0.0.0.0
- GRADIO_SERVER_PORT=7860
- share=True
# - NVIDIA_VISIBLE_DEVICES=0
volumes:
- ./:/app/sensevoice
ports:
- "8000:8000"
runtime: nvidia
deploy:
resources:
reservations:
devices:
- driver: nvidia
# device_ids: ['0']
capabilities: [gpu]
stdin_open: true
tty: true
restart: no
networks:
- sensevoice
entrypoint: ["python", "webui.py", "--listen"]
networks:
sensevoice:
driver: bridge
name: sensevoice
Code sample
https://github.com/IAMJOYBO/sensevoice
Expected behavior
Environment
FROM pytorch/pytorch:2.3.0-cuda12.1-cudnn8-runtime
ENV DEBIAN_FRONTEND=noninteractive
EXPOSE 50000
ENV TZ=Asia/Shanghai
RUN mkdir -p /app
WORKDIR /app
RUN apt update && apt install -y wget curl net-tools tree git git-lfs
RUN git clone https://github.com/FunAudioLLM/SenseVoice.git
WORKDIR /app/SenseVoice
RUN pip install -r requirements.txt
RUN rm -rf /var/lib/apt/lists/*
CMD ["python", "webui.py"]
Additional context
试下启动指令更换为
GRADIO_SERVER_NAME=0.0.0.0 nohup python webui.py > webui.log 2>&1 &
试下启动指令更换为
GRADIO_SERVER_NAME=0.0.0.0 nohup python webui.py > webui.log 2>&1 &
Docker Compose 指定环境变量不可以吗?
官方提供一个镜像?