server: host: 0.0.0.0 port: 80 debug: false data: schema_strict: false recursive: false geoip: enabled: true ip_columns: ["src_ip", "dst_ip"] drop_columns: [] # 默认不清除经纬列——由加载器自动清洗 clustering: algorithm: hdbscan min_cluster_size: 5 random_state: 42 anomaly_detection: enabled: true # PCA采样上限(内存保护) max_sample_rows: 10000 # 风险等级百分位阈值(从数据分布自适应,这里仅做配置覆盖用) risk_levels: critical: 0.99 suspicious: 0.95 watch: 0.80 llm: enabled: true base_url: 'https://api.deepseek.com' api_key: 'sk-ecd33e5341ff4ec9a2c5a5c97a97699b' model: deepseek-v4-flash entity: subnet_masks: [24, 28] ip_columns: ["src_ip", "dst_ip", ":ips", ":ipd", "server-ip", "client-ip"] columns: src_ip: ipv4 dst_ip: ipv4 ":ips": ipv4 ":ipd": ipv4 server-ip: ipv4 client-ip: ipv4 server_ip: ipv4 client_ip: ipv4 bytes_sent: int bytes_rev: int duration: float tls_version: enum cipher_suite: hex dst_url: url lat: lat_lon lon: lat_lon timestamp: timestamp # User column overrides "0ver": enum ":prs": enum ":prd": enum scnt: enum dcnt: enum cnam: string snam: string "4dur": float "8ses": float "2tmo": float "4ksz": int cnrs: enum isrs: enum "8ack": int "8ppk": int "8dbd": timestamp "1ipp": enum "4dbn": enum tabl: string "name": string source-node: string source_node: string ecdhe-named-curve: string ecdhe_named_curve: string "0cph": hex "0crv": hex "0rnd": string "0rnt": float "row": int "time": timestamp ":ips.latd": lat_lon ":ipd.latd": lat_lon ":ips.lond": lat_lon ":ipd.lond": lat_lon ":ips.ispn": string ":ipd.ispn": string ":ips.orgn": string ":ipd.orgn": string ":ips.city": string ":ipd.city": string "+.latd": lat_lon "+.lond": lat_lon "+.ispn": string "+.orgn": string "+.city": string