mirror of
https://github.com/labring/FastGPT.git
synced 2025-07-23 21:13:50 +00:00
V4.6.9-first commit (#899)
* perf: insert mongo dataset data session * perf: dataset data index * remove delay * rename bill schema * rename bill record * perf: bill table * perf: prompt * perf: sub plan * change the usage count * feat: usage bill * publish usages * doc * 新增团队聊天功能 (#20) * perf: doc * feat 添加标签部分 feat 信息团队标签配置 feat 新增团队同步管理 feat team分享页面 feat 完成team分享页面 feat 实现模糊搜索 style 格式化 fix 修复迷糊匹配 style 样式修改 fix 团队标签功能修复 * fix 修复鉴权功能 * merge 合并代码 * fix 修复引用错误 * fix 修复pr问题 * fix 修复ts格式问题 --------- Co-authored-by: archer <545436317@qq.com> Co-authored-by: liuxingwan <liuxingwan.lxw@alibaba-inc.com> * update extra plan * fix: ts * format * perf: bill field * feat: standard plan * fix: ts * feat 个人账号页面修改 (#22) * feat 添加标签部分 feat 信息团队标签配置 feat 新增团队同步管理 feat team分享页面 feat 完成team分享页面 feat 实现模糊搜索 style 格式化 fix 修复迷糊匹配 style 样式修改 fix 团队标签功能修复 * fix 修复鉴权功能 * merge 合并代码 * fix 修复引用错误 * fix 修复pr问题 * fix 修复ts格式问题 * feat 修改个人账号页 --------- Co-authored-by: liuxingwan <liuxingwan.lxw@alibaba-inc.com> * sub plan page (#23) * fix chunk index; error page text * feat: dataset process Integral prediction * feat: stand plan field * feat: sub plan limit * perf: index * query extension * perf: share link push app name * perf: plan point unit * perf: get sub plan * perf: account page * feat 新增套餐详情弹窗代码 (#24) * merge 合并代码 * fix 新增套餐详情弹框 * fix 修复pr问题 * feat: change http node input to prompt editor (#21) * feat: change http node input to prompt editor * fix * split PromptEditor to HttpInput * Team plans (#25) * perf: pay check * perf: team plan test * plan limit check * replace sensitive text * perf: fix some null * collection null check * perf: plans modal * perf: http module * pacakge (#26) * individuation page and pay modal amount (#27) * feat: individuation page * team chat config * pay modal * plan count and replace invalid chars (#29) * fix: user oneapi * fix: training queue * fix: qa queue * perf: remove space chars * replace invalid chars * change httpinput dropdown menu (#28) * perf: http * reseet free plan * perf: plan code to packages * remove llm config to package * perf: code * perf: faq * fix: get team plan --------- Co-authored-by: yst <77910600+yu-and-liu@users.noreply.github.com> Co-authored-by: liuxingwan <liuxingwan.lxw@alibaba-inc.com> Co-authored-by: heheer <71265218+newfish-cmyk@users.noreply.github.com>
This commit is contained in:
Archer
@@ -1,18 +1,19 @@
|
||||
import { replaceVariable } from '@fastgpt/global/common/string/tools';
|
||||
import { getAIApi } from '../config';
|
||||
import { ChatItemType } from '@fastgpt/global/core/chat/type';
|
||||
import { countGptMessagesChars } from '../../chat/utils';
|
||||
|
||||
/*
|
||||
query extension - 问题扩展
|
||||
可以根据上下文，消除指代性问题以及扩展问题，利于检索。
|
||||
*/
|
||||
|
||||
const defaultPrompt = `作为一个向量检索助手，你的任务是结合历史记录，从不同角度，为“原问题”生成个不同版本的“检索词”，从而提高向量检索的语义丰富度，提高向量检索的精度。生成的问题要求指向对象清晰明确。例如：
|
||||
const defaultPrompt = `作为一个向量检索助手，你的任务是结合历史记录，从不同角度，为“原问题”生成个不同版本的“检索词”，从而提高向量检索的语义丰富度，提高向量检索的精度。生成的问题要求指向对象清晰明确，并与原问题语言相同。例如：
|
||||
历史记录:
|
||||
"""
|
||||
"""
|
||||
原问题: 介绍下剧情。
|
||||
检索词: ["发生了什么故事？","故事梗概是什么？","讲述了什么故事？"]
|
||||
检索词: ["介绍下故事的背景和主要人物。","故事的主题是什么？","剧情是是如何发展的？"]
|
||||
----------------
|
||||
历史记录:
|
||||
"""
|
||||
@@ -20,7 +21,7 @@ Q: 对话背景。
|
||||
A: 当前对话是关于 FatGPT 的介绍和使用等。
|
||||
"""
|
||||
原问题: 怎么下载
|
||||
检索词: ["FastGPT 怎么下载？","下载 FastGPT 需要什么条件？","有哪些渠道可以下载 FastGPT？"]
|
||||
检索词: ["FastGPT 如何下载？","下载 FastGPT 需要什么条件？","有哪些渠道可以下载 FastGPT？"]
|
||||
----------------
|
||||
历史记录:
|
||||
"""
|
||||
@@ -30,15 +31,15 @@ Q: 报错 "no connection"
|
||||
A: 报错"no connection"可能是因为……
|
||||
"""
|
||||
原问题: 怎么解决
|
||||
检索词: ["FastGPT 报错"no connection"如何解决？", "报错 'no connection' 是什么原因？", "FastGPT提示'no connection'，要怎么办？"]
|
||||
检索词: ["FastGPT 报错"no connection"如何解决？", "造成 'no connection' 报错的原因。", "FastGPT提示'no connection'，要怎么办？"]
|
||||
----------------
|
||||
历史记录:
|
||||
"""
|
||||
Q: 作者是谁？
|
||||
A: FastGPT 的作者是 labring。
|
||||
"""
|
||||
原问题: 介绍下他
|
||||
检索词: ["介绍下 FastGPT 的作者 labring。","作者 labring 的背景信息。","labring 为什么要做 FastGPT?"]
|
||||
原问题: Tell me about him
|
||||
检索词: ["Introduce labring, the author of FastGPT." ," Background information on author labring." "," Why does labring do FastGPT?"]
|
||||
----------------
|
||||
历史记录:
|
||||
"""
|
||||
@@ -105,8 +106,7 @@ export const queryExtension = async ({
|
||||
rawQuery: string;
|
||||
extensionQueries: string[];
|
||||
model: string;
|
||||
inputTokens: number;
|
||||
outputTokens: number;
|
||||
charsLength: number;
|
||||
}> => {
|
||||
const systemFewShot = chatBg
|
||||
? `Q: 对话背景。
|
||||
@@ -125,18 +125,20 @@ A: ${chatBg}
|
||||
timeout: 480000
|
||||
});
|
||||
|
||||
const messages = [
|
||||
{
|
||||
role: 'user',
|
||||
content: replaceVariable(defaultPrompt, {
|
||||
query: `${query}`,
|
||||
histories: concatFewShot
|
||||
})
|
||||
}
|
||||
];
|
||||
const result = await ai.chat.completions.create({
|
||||
model: model,
|
||||
temperature: 0.01,
|
||||
messages: [
|
||||
{
|
||||
role: 'user',
|
||||
content: replaceVariable(defaultPrompt, {
|
||||
query: `${query}`,
|
||||
histories: concatFewShot
|
||||
})
|
||||
}
|
||||
],
|
||||
// @ts-ignore
|
||||
messages,
|
||||
stream: false
|
||||
});
|
||||
|
||||
@@ -146,8 +148,7 @@ A: ${chatBg}
|
||||
rawQuery: query,
|
||||
extensionQueries: [],
|
||||
model,
|
||||
inputTokens: 0,
|
||||
outputTokens: 0
|
||||
charsLength: 0
|
||||
};
|
||||
}
|
||||
|
||||
@@ -160,8 +161,7 @@ A: ${chatBg}
|
||||
rawQuery: query,
|
||||
extensionQueries: queries,
|
||||
model,
|
||||
inputTokens: result.usage?.prompt_tokens || 0,
|
||||
outputTokens: result.usage?.completion_tokens || 0
|
||||
charsLength: countGptMessagesChars(messages)
|
||||
};
|
||||
} catch (error) {
|
||||
console.log(error);
|
||||
@@ -169,8 +169,7 @@ A: ${chatBg}
|
||||
rawQuery: query,
|
||||
extensionQueries: [],
|
||||
model,
|
||||
inputTokens: 0,
|
||||
outputTokens: 0
|
||||
charsLength: 0
|
||||
};
|
||||
}
|
||||
};
|
||||
|
||||
Reference in New Issue
Block a user