微信扫码
与创始人交个朋友
我要投稿
•智能主动预警:除了支持个人用户以及群聊用户主动搜索进行对话以外,还支持面向m端/pc端指定用户、特定群组,主动发送单聊消息、语音信息(例如定时或固定周期发送报告、识别到的异常信息等)进行提示预警,让问题主动、及时的触达主责用户。
{"metadata": {# 文档级元信息"footers": [], # 页脚"headers": [], # 页眉"catalogs": [] # 目录},"chapters": {# 章节信息"1": "[CHAPTER_ROOT]","1.1": "第一条 xxx","1.2": "第二条 xxxx","1.3": "第三条 xxxx"},"context": [ # 内容信息{# 文本块"text": "JDLxxxx规定","type": "text","pid": 1,"sid": 1,"metadata": {"section_range": []},"cid": "1"},......]}
{"text": [[[0, 0, 1, 1], "名称"],[[0, 1, 1, 2], "尺寸"],[[0, 2, 1, 3], "三层加强材质"],[[0, 3, 1, 4], "售价"],[[0, 4, 1, 5], "三层特硬材质"],[[0, 5, 1, 6], "售价"],[[0, 6, 1, 7], "五层材质"],[[0, 7, 1, 8], "售价"],[[0, 8, 2, 9], "单卷纸生产量"],[[1, 0, 2, 1], "1号纸箱"],[[1, 1, 2, 2], "530*290*370"],[[1, 2, 2, 3], "130/140C/130"],[[1, 3, 2, 4], ""],[[1, 4, 2, 5], "160/160C/160"],[[1, 5, 2, 6], "3.50"],[[1, 6, 2, 7], "140/110B/90/110C/140"],[[1, 7, 2, 8], "3.89"],......],"type": "table","pid": 89,"sid": 111,"metadata": {"section_range": []},"cid": "1.8",}
from langchain import PromptTemplate
from langchain.chains import LLMChain
from langchain.chat_models import ChatOpenAI
def get_condense_question_chain(self):
"""精简问题链"""
CONDENSE_QUESTION_PROMPT = PromptTemplate.from_template(
"""给定历史对话和一个后续问题,将后续问题改写为一个标准问题,用其原始语言,确保避免使用任何不清晰的代词。
历史对话:
{chat_history}
后续输入: {question}
标准问题:"""
)
condense_question_chain = LLMChain(
llm=ChatOpenAI(
model="",
temperature="",
openai_api_key="",
openai_api_base="",
),
prompt=CONDENSE_QUESTION_PROMPT,
)
return condense_question_chain
{
"uid": "742250d1dd9f457aa",
"name": "离线_低装载线路占比_日_3",
"nodes": [
{
"id": "98579cdb14b44423ace0",
"data": {
"viewUid": "e246257e141e4fe78",
"viewSql": "SELECT dt, trans_type_new_name AS trans_type_name , -- 线路类型 transport_org_name, -- 区域 business_type_name, -- 业务类型 team_name, -- 车队 changtu_group, --长途组 low_loading_plink_cnt, plink_cnt FROM bdp_app.app_dis_tsc_product_low_loading_new_sum_d WHERE date_type = 1 AND begin_node_name = '全部' AND add1 = '全部' AND add2 = '全部' AND tail_type = '全部' AND plink = '全部' AND trans_type_old_name = '全部' UNION ALL SELECT dt, trans_type_name, transport_org_name, business_type_name, team_name, changtu_group, low_loading_plink_cnt, plink_cnt FROM bdp_app.app_dis_tsc_product_chuanbai_low_loading_rate_sum_d WHERE date_type = 1"
},
"type": "fact"
}
],
"where": "trans_type_name <> '全部' AND transport_org_name <> '全部' AND business_type_name <> '全部' AND team_name = '全部' AND changtu_group = '全部'",
"measures": [
{
"id": 99,
"names": [
"低装载线路占比"
],
"sql": "SUM(low_loading_plink_cnt)/SUM(plink_cnt)",
"type": "float",
"format": "percentage",
"sort": 1
}
],
"dimensions": [
{
"id": 1,
"names": [
"区域"
],
"field": "transport_org_name",
"type": "str",
"format": "text",
"description": "区域"
},
{
"id": 5,
"names": [
"线路类型"
],
"field": "trans_type_name",
"type": "str",
"format": "text",
"description": "线路类型"
},
{
"id": 12,
"names": [
"业务类型"
],
"field": "business_type_name",
"type": "str",
"format": "text",
"description": "业务类型"
}
],
"timeSeries": [
{
"id": 1,
"names": [
"日",
"日期",
"天"
],
"field": "dt",
"type": "yyyy-mm-dd",
"format": "date"
}
]
"operator": "liuyanchen9",
"updatedAt": 1714112126
}
血缘推理Agent原子能力:
指标
维度&标签
维度值
逻辑模型
视图
物理表
数据分析Agent原子能力
自然语言问询转OLAP
指标波动归因大模型增强分析与解读
扫一扫,加入技术交流群
53AI,企业落地应用大模型首选服务商
产品:大模型应用平台+智能体定制开发+落地咨询服务
承诺:先做场景POC验证,看到效果再签署服务协议。零风险落地应用大模型,已交付160+中大型企业
2024-11-14
关于工作流Chatbot产品的设计思考
2024-11-13
MeetingMind:打造智能会议体验,AI如何简化记录与提升决策
2024-11-11
基于RAGFlow实践Agent——构建智能客服
2024-11-08
用AI做Excel下拉列表,说人话自动生成步骤,太爽了!
2024-11-08
第一批被AI气到崩溃的人出现了
2024-11-08
智能问答系统流程详解:多轮对话与模型训练的技术要点及案例
2024-11-04
利用AI大模型打造客服机器人,传统智能客服可以靠边站了
2024-11-04
浅谈人工智能之基于ollama本地大模型结合本地知识库搭建智能客服
2024-05-08
2024-07-22
2024-07-14
2024-07-11
2024-06-30
2024-03-31
2024-11-08
2024-07-14
2024-08-09
2024-04-01
2024-11-04
2024-10-18
2024-09-27
2024-09-26
2024-09-21
2024-09-17
2024-09-06
2024-09-02