|
|
2d29d8631c
|
fix: pool has already been opened/closed and cannot be reused
|
2024-10-21 10:04:31 +08:00 |
|
|
|
3dc47712e4
|
从数据库中获取风险关键词
|
2024-10-21 10:00:01 +08:00 |
|
|
|
9dc23b714c
|
无风险也更新
|
2024-10-18 19:59:02 +08:00 |
|
|
|
1545a85b09
|
输出风险分析
|
2024-10-18 19:35:26 +08:00 |
|
|
|
115d95bbef
|
udpate README.md
|
2024-10-18 18:38:32 +08:00 |
|
|
|
34ad16ff02
|
dbscan 最大 七天或 10w 数据
|
2024-10-18 18:10:23 +08:00 |
|
|
|
8a6db8f8f2
|
add requirements_version.txt
|
2024-10-18 17:26:48 +08:00 |
|
|
|
cae3877048
|
risk-analyze main 循环
|
2024-10-18 16:59:43 +08:00 |
|
|
|
3d36dcadf6
|
add README.md
|
2024-10-18 16:46:19 +08:00 |
|
|
|
a051674f2e
|
feat: 分析结果入mysql
|
2024-10-18 16:01:29 +08:00 |
|
|
|
47ae4dc8d5
|
fix: 二维task文档匹配问题
|
2024-10-18 15:47:18 +08:00 |
|
|
|
d52f37e5f0
|
暂存
二位数组转一维,文档ID标记的问题还没解决
|
2024-10-18 15:43:37 +08:00 |
|
|
|
ad3ef8e504
|
添加 kmeans (存档)
|
2024-10-10 18:31:47 +08:00 |
|
|
|
b3fa8edf60
|
添加 cmd.risk-analyze
|
2024-10-10 18:31:35 +08:00 |
|
|
|
004552f4d5
|
添加dbscan
|
2024-10-10 18:19:50 +08:00 |
|
|
|
a49db1b71b
|
使用 title or content 作为embedding内容
|
2024-10-08 14:35:00 +08:00 |
|
|
|
16f4469365
|
调整embedding策略
只处理2周内的新闻,利用索引、时间降序
|
2024-09-23 15:13:33 +08:00 |
|
|
|
341435e603
|
WIP: LLM 风险判断
|
2024-09-20 18:33:56 +08:00 |
|
|
|
fff2a32d7e
|
从ES获取的结果中过滤掉\x00
|
2024-09-20 15:20:19 +08:00 |
|
|
|
43ddd8665c
|
fix: ES 同步接口时间精确到秒
|
2024-09-20 11:44:04 +08:00 |
|
|
|
1e1f92461a
|
typo
|
2024-09-19 14:35:34 +08:00 |
|
|
|
71fd688227
|
修改 OPENAI 环境变量为 OPENAI_EMBEDDING
|
2024-09-19 14:29:30 +08:00 |
|
|
|
0b658dee88
|
添加 embedding 错误处理
|
2024-09-19 14:27:42 +08:00 |
|
|
|
92e3699cd8
|
添加 cl100k_base 字典与 text-embedding-3 适配
|
2024-09-19 14:26:20 +08:00 |
|
|
|
4b5e59f35d
|
同步添加重试机制
|
2024-09-19 11:07:16 +08:00 |
|
|
|
7395d98ce3
|
延迟一小时入库
-1 hour -> start_time 相当于即时入库
-2 hour -> start_time -1 -> end_time 相当于延迟一小时入库
|
2024-09-19 10:55:01 +08:00 |
|
|
|
dd41d6fa5f
|
fix: es 返回空数据
|
2024-09-14 17:19:39 +08:00 |
|
|
|
07afdf21b8
|
使用清华源镜像
|
2024-09-14 17:01:16 +08:00 |
|
|
|
698f3fd65e
|
设置时区 TZ=Asia/Shanghai
|
2024-09-14 17:00:50 +08:00 |
|
|
|
74292e91e7
|
添加 print 函数
|
2024-09-14 16:51:51 +08:00 |
|
|
|
bc0fffc079
|
fix: 无限递归
|
2024-09-14 16:36:19 +08:00 |
|
|
|
615d96c663
|
Fix: es_intervals 为空
|
2024-09-14 12:00:10 +08:00 |
|
|
|
2d5eae2a18
|
添加dockerfile
|
2024-09-14 11:59:56 +08:00 |
|
|
|
1198df2d4f
|
删除tiktoken依赖
|
2024-09-14 10:27:54 +08:00 |
|
|
|
48d010007a
|
循环更新 embedding
|
2024-09-14 10:22:32 +08:00 |
|
|
|
b2ed1394e0
|
增加刷新 openai embedding 功能
|
2024-09-11 18:41:44 +08:00 |
|
|
|
09b22517df
|
format
|
2024-09-11 16:25:17 +08:00 |
|
|
|
c9d361f56b
|
从 pg info 字段中提取出需要的字段
|
2024-09-11 14:48:36 +08:00 |
|
|
|
8009a8295c
|
实现 同步数据到 mysql
|
2024-09-11 14:23:48 +08:00 |
|
|
|
5794c8d5c5
|
first commit
|
2024-09-11 09:32:12 +08:00 |
|