-
Notifications
You must be signed in to change notification settings - Fork 10
Expand file tree
/
Copy pathpeople-worker.yaml.example
More file actions
53 lines (48 loc) · 1.34 KB
/
Copy pathpeople-worker.yaml.example
File metadata and controls
53 lines (48 loc) · 1.34 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
# relive-people-worker 配置示例
# 用于 Mac M4 等高性能设备离线执行人脸检测
#
# 使用方法:
# 1. 复制此文件: cp people-worker.yaml.example people-worker.yaml
# 2. 编辑配置,设置正确的 endpoint 和 api_key
# 3. 运行检查: ./relive-people-worker check -config people-worker.yaml
# 4. 启动 Worker: ./relive-people-worker run -config people-worker.yaml
server:
# NAS 服务端地址
endpoint: "http://nas:8080"
# API Key(用于认证)
# 可以从环境变量读取: "${RELIVE_API_KEY}"
api_key: "your-api-key-here"
# API 请求超时(秒)
timeout: 30
people_worker:
# Worker 实例ID(用于标识,可选)
worker_id: "mac-m4"
# 并发处理数(M4 Mac 推荐 4-8)
workers: 4
# 每批获取任务数
fetch_limit: 10
# 重试次数
retry_count: 3
# 重试延迟(秒)
retry_delay: 5
ml:
# relive-ml 服务地址(本地运行)
endpoint: "http://localhost:5050"
# 人脸检测超时(秒)
timeout: 15
download:
# 临时文件目录
temp_dir: "~/.relive-people-worker/temp"
# 下载超时(秒)
timeout: 30
# 最大并发下载
max_concurrent: 4
# 下载重试次数
retry_count: 3
logging:
# 日志级别: debug, info, warn, error
level: "info"
# 输出到控制台
console: true
# 日志文件路径(空表示不写入文件)
file: ""