lixiaolong
b386f003a0
- 集成Qwen3.8-27B-FP8快速模型和Qwen3.8-27B精确模型作为SGLang服务 - 添加SGLang API配置选项(SGLANG_FAST_BASE_URL、SGLANG_ACCURATE_BASE_URL等) - 实现音频分类中的决策聚合算法(topK、nearestWeight、labelMaxDistance) - 添加视频采样帧限制(VIDEO_SAMPLING_FRAME_LIMIT)和上下文token限制 - 更新健康检查以监控SGLang服务状态 - 实现视频分析的双模式决策策略(快速+精确) - 添加音频参考文件导入工具(import_audio_references.py) - 扩展音频分类标签支持FIND_VEHICLE_HORN类别 - 优化视频分析的帧采样策略，始终包含视频尾部帧 - 添加决策策略参数(tuning、decisionPolicy)支持 - 更新配置类以支持新的SGLang和视频参数 - 修改compose配置以支持Qwen3.8模型部署 - 更新音频分类测试用例验证聚合逻辑 - 重构视频测试以支持SGLang API格式和决策策略
44 lines
1.2 KiB
YAML
44 lines
1.2 KiB
YAML
name: cmvr-ai-analysis-qwen38
|
|
|
|
services:
|
|
analysis-service:
|
|
container_name: cmvr-ai-analysis
|
|
image: cmvr-ai-analysis:1.2.0-qwen38
|
|
build:
|
|
context: ..
|
|
dockerfile: Dockerfile
|
|
restart: unless-stopped
|
|
ports:
|
|
- "192.168.28.10:14080:8080"
|
|
env_file:
|
|
- .env
|
|
environment:
|
|
TZ: Asia/Shanghai
|
|
PROFILES_DIR: /data/profiles
|
|
ARTIFACTS_DIR: /data/artifacts
|
|
JOBS_DIR: /data/jobs
|
|
OLLAMA_BASE_URL: http://host.docker.internal:11434
|
|
OLLAMA_MODEL: qwen3-vl:32b
|
|
SGLANG_FAST_BASE_URL: http://192.168.28.10:14081/v1
|
|
SGLANG_ACCURATE_BASE_URL: http://192.168.28.10:14082/v1
|
|
SGLANG_TIMEOUT_SECONDS: "600"
|
|
VISION_OLLAMA_FALLBACK_ENABLED: "true"
|
|
extra_hosts:
|
|
- "host.docker.internal:host-gateway"
|
|
volumes:
|
|
- ../data/profiles:/data/profiles:ro
|
|
- ../data/artifacts:/data/artifacts
|
|
- ../data/jobs:/data/jobs
|
|
- ../data/logs:/data/logs
|
|
healthcheck:
|
|
test: ["CMD", "curl", "-fsS", "http://127.0.0.1:8080/health"]
|
|
interval: 30s
|
|
timeout: 5s
|
|
retries: 3
|
|
start_period: 20s
|
|
logging:
|
|
driver: json-file
|
|
options:
|
|
max-size: "100m"
|
|
max-file: "5"
|