diff --git a/content/LangChain for LLM Application Development/1.开篇介绍.md b/content/LangChain for LLM Application Development/1.开篇介绍.md deleted file mode 100644 index 67e9cbd..0000000 --- a/content/LangChain for LLM Application Development/1.开篇介绍.md +++ /dev/null @@ -1,33 +0,0 @@ -## 吴恩达 LangChain大模型应用开发 开端篇 - -## LangChain for LLM Application Development - -欢迎来到LangChain大模型应用开发短期课程👏🏻👏🏻 - -本课程由哈里森·蔡斯 (Harrison Chase，LangChain作者）与Deeplearning.ai合作开发，旨在教大家使用这个神奇工具。 - -### 🚀 LangChain的诞生和发展 - -通过对LLM或大型语言模型给出提示(prompt)，现在可以比以往更快地开发AI应用程序，但是一个应用程序可能需要进行多轮提示以及解析输出。 - -在此过程有很多胶水代码需要编写，基于此需求，哈里森·蔡斯 (Harrison Chase) 创建了LangChain，使开发过程变得更加丝滑。 - -LangChain开源社区快速发展，贡献者已达数百人，正以惊人的速度更新代码和功能。 - - -### 📚 课程基本内容 - -LangChain是用于构建大模型应用程序的开源框架，有Python和JavaScript两个不同版本的包。LangChain基于模块化组合，有许多单独的组件，可以一起使用或单独使用。此外LangChain还拥有很多应用案例，帮助我们了解如何将这些模块化组件以链式方式组合，以形成更多端到端的应用程序 。 - -在本课程中，我们将介绍LandChain的常见组件。具体而言我们会讨论一下几个方面 -- 模型(Models) -- 提示(Prompts): 使模型执行操作的方式 -- 索引(Indexes): 获取数据的方式，可以与模型结合使用 -- 链式(Chains): 端到端功能实现 -- 代理(Agents): 使用模型作为推理引擎 - - - -### 🌹致谢课程重要贡献者 - -最后特别感谢Ankush Gholar（LandChain的联合作者）、Geoff Ladwig,、Eddy Shyu 以及 Diala Ezzedine，他们也为本课程内容贡献颇多~ \ No newline at end of file diff --git a/content/LangChain for LLM Application Development/1.简介.ipynb b/content/LangChain for LLM Application Development/1.简介.ipynb new file mode 100644 index 0000000..62cee6b --- /dev/null +++ b/content/LangChain for LLM Application Development/1.简介.ipynb @@ -0,0 +1,87 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "cfab521b-77fa-41be-a964-1f50f2ef4689", + "metadata": {}, + "source": [ + "# 1. 简介\n", + "
\n", + "\n", + "欢迎来到LangChain大模型应用开发短期课程👏🏻👏🏻\n", + "\n", + "本课程由哈里森·蔡斯 (Harrison Chase，LangChain作者）与Deeplearning.ai合作开发，旨在教大家使用这个神奇工具。\n", + "\n", + "\n", + "\n", + "## 1.1 LangChain的诞生和发展\n", + "\n", + "通过对LLM或大型语言模型给出提示(prompt)，现在可以比以往更快地开发AI应用程序，但是一个应用程序可能需要进行多轮提示以及解析输出。\n", + "\n", + "在此过程有很多胶水代码需要编写，基于此需求，哈里森·蔡斯 (Harrison Chase) 创建了LangChain，使开发过程变得更加丝滑。\n", + "\n", + "LangChain开源社区快速发展，贡献者已达数百人，正以惊人的速度更新代码和功能。\n", + "\n", + "\n", + "## 1.2 课程基本内容\n", + "\n", + "LangChain是用于构建大模型应用程序的开源框架，有Python和JavaScript两个不同版本的包。LangChain基于模块化组合，有许多单独的组件，可以一起使用或单独使用。此外LangChain还拥有很多应用案例，帮助我们了解如何将这些模块化组件以链式方式组合，以形成更多端到端的应用程序 。\n", + "\n", + "在本课程中，我们将介绍LandChain的常见组件。具体而言我们会讨论一下几个方面\n", + "- 模型(Models)\n", + "- 提示(Prompts): 使模型执行操作的方式\n", + "- 索引(Indexes): 获取数据的方式，可以与模型结合使用\n", + "- 链式(Chains): 端到端功能实现\n", + "- 代理(Agents): 使用模型作为推理引擎\n", + "\n", + " \n", + "\n", + "## 1.3 致谢课程重要贡献者\n", + "\n", + "最后特别感谢Ankush Gholar（LandChain的联合作者）、Geoff Ladwig,、Eddy Shyu 以及 Diala Ezzedine，他们也为本课程内容贡献颇多~ " + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e3618ca8", + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/content/LangChain for LLM Application Development/2.模型、提示和解析器.ipynb b/content/LangChain for LLM Application Development/2.模型、提示和解析器.ipynb index d06b746..35a85ea 100644 --- a/content/LangChain for LLM Application Development/2.模型、提示和解析器.ipynb +++ b/content/LangChain for LLM Application Development/2.模型、提示和解析器.ipynb @@ -1,38 +1,58 @@ { "cells": [ { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ - "# 模型，提示和输出解释器\n", + "# 2. 模型，提示和输出解释器\n", "\n", - "\n", - "**目录**\n", - "* 获取你的OpenAI API Key\n", - "* 直接调用OpenAI的API\n", - "* 通过LangChain进行的API调用：\n", - " * 提示（Prompts）\n", - " * [模型（Models)](#model)\n", - " * 输出解析器（Output parsers）\n", - " " + "\n", + "\n", + "\n", + "" ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": { "tags": [] }, "source": [ - "## 获取你的OpenAI API Key\n", + "## 2.1 获取你的OpenAI API Key\n", "\n", "登陆[OpenAI账户获取你的API Key](https://platform.openai.com/account/api-keys) " ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, "outputs": [], "source": [ @@ -60,13 +80,12 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": { "tags": [] }, "source": [ - "## Chat API：OpenAI\n", + "## 2.2 使用Chat API：OpenAI\n", "\n", "我们先从直接调用OpenAI的API开始。\n", "\n", @@ -99,13 +118,12 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": { "tags": [] }, "source": [ - "### 一个简单的例子\n", + "### 2.2.1 一个简单的例子\n", "\n", "我们来一个简单的例子 - 分别用中英文问问模型\n", "\n", @@ -156,11 +174,10 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ - "### 复杂一点例子\n", + "### 2.2.2 复杂一点的例子\n", "\n", "上面的简单例子，模型`gpt-3.5-turbo`对我们的关于1+1是什么的提问给出了回答。\n", "\n", @@ -188,7 +205,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -215,7 +231,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -255,7 +270,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -296,7 +310,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -310,13 +323,12 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": { "tags": [] }, "source": [ - "## Chat API：LangChain\n", + "## 2.3 使用Chat API：LangChain\n", "\n", "在前面一部分，我们通过封装函数`get_completion`直接调用了OpenAI完成了对海岛邮件进行了翻译，得到用平和尊重的语气、美式英语表达的邮件。\n", "\n", @@ -337,13 +349,12 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": { "tags": [] }, "source": [ - "### 模型\n", + "### 2.3.1 模型\n", "\n", "从`langchain.chat_models`导入`OpenAI`的对话模型`ChatOpenAI`。 除去OpenAI以外，`langchain.chat_models`还集成了其他对话模型，更多细节可以查看[Langchain官方文档](https://python.langchain.com/en/latest/modules/models/chat/integrations.html)。" ] @@ -385,7 +396,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -393,11 +403,10 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ - "### 提示模板\n", + "### 2.3.2 提示模板\n", "\n", "在前面的例子中，我们通过[f字符串](https://docs.python.org/zh-cn/3/tutorial/inputoutput.html#tut-f-strings)把Python表达式的值`style`和`customer_email`添加到`prompt`字符串内。\n", "\n", @@ -412,7 +421,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -437,7 +445,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -481,7 +488,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -511,7 +517,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -554,7 +559,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -596,7 +600,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -625,7 +628,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -665,7 +667,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -735,7 +736,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -765,7 +765,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": { "tags": [] @@ -777,7 +776,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -861,15 +859,13 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ - "### 输出解析器" + "### 2.3.3 输出解析器" ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -906,7 +902,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -941,7 +936,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -968,7 +962,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -985,7 +978,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1016,7 +1008,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1066,7 +1057,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1074,7 +1064,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1106,7 +1095,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1123,7 +1111,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1183,7 +1170,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1237,7 +1223,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1269,7 +1254,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1300,7 +1284,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1349,15 +1332,13 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ - "## 补充材料" + "## 2.4 补充材料" ] }, { - "attachments": {}, "cell_type": "markdown", "metadata": {}, "source": [ @@ -1435,8 +1416,30 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.11" - } + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": { + "height": "calc(100% - 180px)", + "left": "10px", + "top": "150px", + "width": "165px" + }, + "toc_section_display": true, + "toc_window_display": true + }, + "toc-autonumbering": false, + "toc-showcode": false, + "toc-showmarkdowntxt": false, + "toc-showtags": false }, "nbformat": 4, "nbformat_minor": 4 diff --git a/content/LangChain for LLM Application Development/3.存储 .ipynb b/content/LangChain for LLM Application Development/3.存储.ipynb similarity index 97% rename from content/LangChain for LLM Application Development/3.存储 .ipynb rename to content/LangChain for LLM Application Development/3.存储.ipynb index 72f5059..305bc33 100644 --- a/content/LangChain for LLM Application Development/3.存储 .ipynb +++ b/content/LangChain for LLM Application Development/3.存储.ipynb @@ -1,12 +1,14 @@ { "cells": [ { - "attachments": {}, "cell_type": "markdown", "id": "a786c77c", "metadata": {}, "source": [ - "# LangChain: Memory（记忆）\n", + "# 3. 储存\n", + "\n", + "
\n", + "\n", "当你与那些语言模型进行交互的时候，他们不会记得你之前和他进行的交流内容，这在我们构建一些应用程序（如聊天机器人）的时候，是一个很大的问题---显得不够智能！\n", "\n", "因此，在本节中我们将介绍LangChain 中的 **Memory(记忆)** 模块，即他是如何将先前的对话嵌入到语言模型中的，使其具有连续对话的能力\n", @@ -29,7 +31,7 @@ " \n", "此次课程主要介绍其中四种记忆模块，其他模块可查看文档学习。\n", "\n", - "## 大纲\n", + "\n", "* ConversationBufferMemory（对话缓存记忆）\n", "* ConversationBufferWindowMemory（对话缓存窗口记忆）\n", "* ConversationTokenBufferMemory（对话令牌缓存记忆）\n", @@ -37,7 +39,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7e10db6f", "metadata": {}, @@ -48,18 +49,16 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1297dcd5", "metadata": {}, "source": [ - "## ConversationBufferMemory（对话缓存记忆） \n", + "## 3.1 对话缓存储存 \n", " \n", "这种记忆允许存储消息，然后从变量中提取消息。" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "0768ca9b", "metadata": {}, @@ -68,7 +67,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "d76f6ba7", "metadata": {}, @@ -133,7 +131,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "dea83837", "metadata": {}, @@ -142,7 +139,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1a3b4c42", "metadata": {}, @@ -232,7 +228,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "e71564ad", "metadata": {}, @@ -241,7 +236,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "54d006bd", "metadata": {}, @@ -345,7 +339,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "33cb734b", "metadata": {}, @@ -354,7 +347,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "0393df3d", "metadata": {}, @@ -462,7 +454,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "5a96a8d9", "metadata": {}, @@ -532,7 +523,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "6bd222c3", "metadata": {}, @@ -541,7 +531,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "0b5de846", "metadata": {}, @@ -595,7 +584,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "07d2e892", "metadata": {}, @@ -698,7 +686,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "2ac544f2", "metadata": {}, @@ -767,7 +754,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "8839314a", "metadata": {}, @@ -778,12 +764,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "cf98e9ff", "metadata": {}, "source": [ - "## ConversationBufferWindowMemory（对话缓存窗口记忆）\n", + "## 3.2 对话缓存窗口储存\n", " \n", "随着对话变得越来越长，所需的内存量也变得非常长。将大量的tokens发送到LLM的成本，也会变得更加昂贵,这也就是为什么API的调用费用，通常是基于它需要处理的tokens数量而收费的。\n", " \n", @@ -804,7 +789,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "641477a4", "metadata": {}, @@ -863,7 +847,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "9b401f0b", "metadata": {}, @@ -899,7 +882,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "63bda148", "metadata": {}, @@ -927,7 +909,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "b6d661e3", "metadata": {}, @@ -1005,7 +986,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "a1080168", "metadata": {}, @@ -1038,16 +1018,14 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "d2931b92", "metadata": {}, "source": [ - "## ConversationTokenBufferMemory（对话token缓存记忆）" + "## 3.3 对话token缓存储存" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "dff5b4c7", "metadata": {}, @@ -1069,7 +1047,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "2187cfe6", "metadata": {}, @@ -1094,7 +1071,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "f3a84112", "metadata": {}, @@ -1121,7 +1097,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7b62b2e1", "metadata": {}, @@ -1153,7 +1128,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "f7f6be43", "metadata": {}, @@ -1188,7 +1162,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "5e4d918b", "metadata": {}, @@ -1204,16 +1177,14 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "5ff55d5d", "metadata": {}, "source": [ - "## ConversationSummaryBufferMemory（对话摘要缓存记忆）" + "## 3.4 对话摘要缓存储存" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7d39b83a", "metadata": {}, @@ -1247,7 +1218,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "6572ef39", "metadata": {}, @@ -1305,7 +1275,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7ccb97b6", "metadata": {}, @@ -1396,7 +1365,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "4ba827aa", "metadata": { @@ -1530,7 +1498,20 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.9" + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true } }, "nbformat": 4, diff --git a/content/LangChain for LLM Application Development/4.模型链.ipynb b/content/LangChain for LLM Application Development/4.模型链.ipynb index 4eadf48..396fe07 100644 --- a/content/LangChain for LLM Application Development/4.模型链.ipynb +++ b/content/LangChain for LLM Application Development/4.模型链.ipynb @@ -1,30 +1,29 @@ { "cells": [ { - "attachments": {}, + "cell_type": "markdown", + "id": "7ee04154", + "metadata": { + "toc": true + }, + "source": [ + "
  • 4. 模型链
    • " + ] + }, + { "cell_type": "markdown", "id": "52824b89-532a-4e54-87e9-1410813cd39e", "metadata": {}, "source": [ - "# Chains in LangChain（LangChain中的链）\n", - "\n", - "## Outline\n", - "\n", - "* LLMChain（大语言模型链）\n", - "* Sequential Chains（顺序链）\n", - " * SimpleSequentialChain\n", - " * SequentialChain\n", - "* Router Chain（路由链）" + "# 4. 模型链" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "54810ef7", "metadata": {}, "source": [ - "### 为什么我们需要Chains ？\n", - "链允许我们将多个组件组合在一起，以创建一个单一的、连贯的应用程序。链（Chains）通常将一个LLM（大语言模型）与提示结合在一起，使用这个构建块，您还可以将一堆这些构建块组合在一起，对您的文本或其他数据进行一系列操作。例如，我们可以创建一个链，该链接受用户输入，使用提示模板对其进行格式化，然后将格式化的响应传递给LLM。我们可以通过将多个链组合在一起，或者通过将链与其他组件组合在一起来构建更复杂的链。" + "模型链允许我们将多个组件组合在一起，以创建一个单一的、连贯的应用程序。链（Chains）通常将一个LLM（大语言模型）与提示结合在一起，使用这个构建块，您还可以将一堆这些构建块组合在一起，对您的文本或其他数据进行一系列操作。例如，我们可以创建一个链，该链接受用户输入，使用提示模板对其进行格式化，然后将格式化的响应传递给LLM。我们可以通过将多个链组合在一起，或者通过将链与其他组件组合在一起来构建更复杂的链。" ] }, { @@ -64,7 +63,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "663fc885", "metadata": {}, @@ -163,21 +161,19 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "b940ce7c", "metadata": {}, "source": [ - "## 1. LLMChain" + "## 4.1 大语言模型链" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "e000bd16", "metadata": {}, "source": [ - "LLMChain是一个简单但非常强大的链，也是后面我们将要介绍的许多链的基础。" + "大语言模型链(LLMChain)是一个简单但非常强大的链，也是后面我们将要介绍的许多链的基础。" ] }, { @@ -193,7 +189,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "94a32c6f", "metadata": {}, @@ -213,7 +208,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "81887434", "metadata": {}, @@ -235,7 +229,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "5c22cb13", "metadata": {}, @@ -254,7 +247,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "8d7d5ff6", "metadata": {}, @@ -285,7 +277,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1e1ede1c", "metadata": {}, @@ -321,21 +312,19 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "49158430", "metadata": {}, "source": [ - "## 2. Sequential Chains" + "## 4.2 顺序链" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "69b03469", "metadata": {}, "source": [ - "### 2.1 SimpleSequentialChain\n", + "### 4.2.1 简单顺序链\n", "\n", "顺序链（Sequential Chains）是按预定义顺序执行其链接的链。具体来说，我们将使用简单顺序链（SimpleSequentialChain），这是顺序链的最简单类型，其中每个步骤都有一个输入/输出，一个步骤的输出是下一个步骤的输入" ] @@ -362,7 +351,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "0e732589", "metadata": {}, @@ -389,7 +377,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "dcfca7bd", "metadata": {}, @@ -415,7 +402,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "3a1991f4", "metadata": {}, @@ -436,7 +422,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "5122f26a", "metadata": {}, @@ -530,16 +515,14 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7b5ce18c", "metadata": {}, "source": [ - "### 2.2 SequentialChain" + "### 4.2.2 顺序链" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "1e69f4c0", "metadata": {}, @@ -563,7 +546,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "3d4be4e8", "metadata": {}, @@ -583,7 +565,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "9811445c", "metadata": {}, @@ -690,7 +671,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "0509de01", "metadata": {}, @@ -751,16 +731,14 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "3041ea4c", "metadata": {}, "source": [ - "## 3. Router Chain（路由链）" + "## 4.3. 路由链" ] }, { - "attachments": {}, "cell_type": "markdown", "id": "f0c32f97", "metadata": {}, @@ -769,7 +747,7 @@ "\n", "一个相当常见但基本的操作是根据输入将其路由到一条链，具体取决于该输入到底是什么。如果你有多个子链，每个子链都专门用于特定类型的输入，那么可以组成一个路由链，它首先决定将它传递给哪个子链，然后将它传递给那个链。\n", "\n", - "路由器由两个组件组成：\n", + "Router Chain（路由链）由两个组件组成：\n", "\n", "- 路由器链本身（负责选择要调用的下一个链）\n", "- destination_chains：路由器链可以路由到的链\n", @@ -778,12 +756,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "cb1b4708", "metadata": {}, "source": [ - "### 定义提示模板" + "### 4.3.1 定义提示模板" ] }, { @@ -846,7 +823,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "6922b35e", "metadata": {}, @@ -886,12 +862,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "80eb1de8", "metadata": {}, "source": [ - "### 导入相关的包" + "### 4.3.2 导入相关的包" ] }, { @@ -907,12 +882,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "50c16f01", "metadata": {}, "source": [ - "### 定义语言模型" + "### 4.3.3 定义语言模型" ] }, { @@ -927,12 +901,13 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "8795cd42", "metadata": {}, "source": [ - "### LLMRouterChain（此链使用 LLM 来确定如何路由事物）\n", + "### 4.3.4 大语言模型路由链\n", + "\n", + "大语言模型路由链（LLMRouterChain）使用大语言模型（LLM）来确定如何路由事物\n", "\n", "在这里，我们需要一个**多提示链**。这是一种特定类型的链，用于在多个不同的提示模板之间进行路由。\n", "但是，这只是你可以路由的一种类型。你也可以在任何类型的链之间进行路由。\n", @@ -942,7 +917,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "46633b43", "metadata": {}, @@ -972,7 +946,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "eba115de", "metadata": {}, @@ -993,7 +966,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "948700c4", "metadata": {}, @@ -1041,7 +1013,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "de5c46d0", "metadata": {}, @@ -1075,7 +1046,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7e92355c", "metadata": {}, @@ -1099,7 +1069,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "086503f7", "metadata": {}, @@ -1108,7 +1077,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "969cd878", "metadata": {}, @@ -1183,7 +1151,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "289c5ca9", "metadata": {}, @@ -1225,7 +1192,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "4186a2b9", "metadata": {}, @@ -1318,7 +1284,25 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.9.16" + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": true, + "toc_position": { + "height": "calc(100% - 180px)", + "left": "10px", + "top": "150px", + "width": "261.818px" + }, + "toc_section_display": true, + "toc_window_display": true } }, "nbformat": 4, diff --git a/content/LangChain for LLM Application Development/5.文档问答.ipynb b/content/LangChain for LLM Application Development/5.基于文档的问答.ipynb similarity index 94% rename from content/LangChain for LLM Application Development/5.文档问答.ipynb rename to content/LangChain for LLM Application Development/5.基于文档的问答.ipynb index 6993bc7..811e8df 100644 --- a/content/LangChain for LLM Application Development/5.文档问答.ipynb +++ b/content/LangChain for LLM Application Development/5.基于文档的问答.ipynb @@ -1,11 +1,20 @@ { "cells": [ + { + "cell_type": "markdown", + "id": "f200ba9a", + "metadata": {}, + "source": [ + "# 5 基于文档的问答 \n", + "
    " + ] + }, { "cell_type": "markdown", "id": "52824b89-532a-4e54-87e9-1410813cd39e", "metadata": {}, "source": [ - "# 第四章 基于LangChain的文档问答\n", + "\n", "本章内容主要利用langchain构建向量数据库，可以在文档上方或关于文档回答问题，因此，给定从PDF文件、网页或某些公司的内部文档收集中提取的文本，使用llm回答有关这些文档内容的问题" ] }, @@ -16,18 +25,22 @@ "height": 30 }, "source": [ - "## 环境配置\n", + "\n", "\n", "安装langchain，设置chatGPT的OPENAI_API_KEY\n", + "\n", "* 安装langchain\n", + "\n", "```\n", "pip install langchain\n", "```\n", "* 安装docarray\n", + "\n", "```\n", "pip install docarray\n", "```\n", "* 设置API-KEY环境变量\n", + "\n", "```\n", "export OPENAI_API_KEY='api-key'\n", "\n", @@ -81,7 +94,7 @@ "height": 30 }, "source": [ - "### 导入embedding模型和向量存储组件\n", + "## 5.1 导入embedding模型和向量存储组件\n", "使用Dock Array内存搜索向量存储，作为一个内存向量存储，不需要连接外部数据库" ] }, @@ -264,7 +277,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "3bd6422c", "metadata": {}, @@ -273,12 +285,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "2963fc63", "metadata": {}, "source": [ - "#### 创建向量存储\n", + "### 5.1.2 创建向量存储\n", "将导入一个索引，即向量存储索引创建器" ] }, @@ -373,7 +384,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "eb74cc79", "metadata": {}, @@ -382,12 +392,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "dd34e50e", "metadata": {}, "source": [ - "#### 使用语言模型与文档结合使用\n", + "### 5.1.3 使用语言模型与文档结合使用\n", "想要使用语言模型并将其与我们的许多文档结合使用，但是语言模型一次只能检查几千个单词，如果我们有非常大的文档，如何让语言模型回答关于其中所有内容的问题呢？通过embedding和向量存储实现\n", "* embedding \n", "文本片段创建数值表示文本语义，相似内容的文本片段将具有相似的向量，这使我们可以在向量空间中比较文本片段\n", @@ -594,12 +603,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "fe41b36f", "metadata": {}, "source": [ - "### 如何回答我们文档的相关问题\n", + "## 5.2 如何回答我们文档的相关问题\n", "首先，我们需要从这个向量存储中创建一个检索器，检索器是一个通用接口，可以由任何接受查询并返回文档的方法支持。接下来，因为我们想要进行文本生成并返回自然语言响应\n" ] }, @@ -685,7 +693,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "12f042e7", "metadata": {}, @@ -779,7 +786,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "e28c5657", "metadata": {}, @@ -788,12 +794,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "44f1fa38", "metadata": {}, "source": [ - "#### 不同类型的chain链\n", + "### 5.2.1 不同类型的chain链\n", "想在许多不同类型的块上执行相同类型的问答，该怎么办？之前的实验中只返回了4个文档，如果有多个文档，那么我们可以使用几种不同的方法\n", "* Map Reduce \n", "将所有块与问题一起传递给语言模型，获取回复，使用另一个语言模型调用将所有单独的回复总结成最终答案，它可以在任意数量的文档上运行。可以并行处理单个问题，同时也需要更多的调用。它将所有文档视为独立的\n", @@ -804,12 +809,6 @@ "* Stuff \n", "将所有内容组合成一个文档" ] - }, - { - "cell_type": "markdown", - "id": "7988f412", - "metadata": {}, - "source": [] } ], "metadata": { @@ -828,7 +827,20 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.9.16" + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true } }, "nbformat": 4, diff --git a/content/LangChain for LLM Application Development/6.评估.ipynb b/content/LangChain for LLM Application Development/6.评估.ipynb index 081a749..1125634 100644 --- a/content/LangChain for LLM Application Development/6.评估.ipynb +++ b/content/LangChain for LLM Application Development/6.评估.ipynb @@ -1,12 +1,14 @@ { "cells": [ { - "attachments": {}, "cell_type": "markdown", "id": "52824b89-532a-4e54-87e9-1410813cd39e", "metadata": {}, "source": [ - "# 第五章 如何评估基于LLM的应用程序\n", + "# 6. 评估\n", + "\n", + "
    \n", + "\n", "当使用llm构建复杂应用程序时，评估应用程序的表现是一个重要但有时棘手的步骤，它是否满足某些准确性标准？\n", "通常更有用的是从许多不同的数据点中获得更全面的模型表现情况\n", "一种是使用语言模型本身和链本身来评估其他语言模型、其他链和其他应用程序" @@ -29,12 +31,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "28008949", "metadata": {}, "source": [ - "## 创建LLM应用\n", + "## 6.1 创建LLM应用\n", "按照langchain链的方式进行构建" ] }, @@ -255,12 +256,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "791ebd73", "metadata": {}, "source": [ - "### 创建评估数据点\n", + "### 6.1.1 创建评估数据点\n", "们需要做的第一件事是真正弄清楚我们想要评估它的一些数据点，我们将介绍几种不同的方法来完成这个任务\n", "1、将自己想出好的数据点作为例子，查看一些数据，然后想出例子问题和答案，以便以后用于评估" ] @@ -312,7 +312,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "b9c52116", "metadata": {}, @@ -321,12 +320,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "8d548aef", "metadata": {}, "source": [ - "### 创建测试用例数据\n" + "### 6.1.2 创建测试用例数据\n" ] }, { @@ -353,7 +351,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "b73ce510", "metadata": {}, @@ -363,12 +360,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "c7ce3e4f", "metadata": {}, "source": [ - "### 通过LLM生成测试用例" + "### 6.1.3 通过LLM生成测试用例" ] }, { @@ -487,12 +483,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "faf25f2f", "metadata": {}, "source": [ - "### 组合用例数据" + "### 6.1.4 组合用例数据" ] }, { @@ -542,12 +537,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "63f3cb08", "metadata": {}, "source": [ - "## 人工评估\n", + "## 6.2 人工评估\n", "现在有了这些示例，但是我们如何评估正在发生的事情呢？\n", "通过运行一个示例通过链，并查看它产生的输出\n", "在这里我们传递一个查询，然后我们得到一个答案。实际上正在发生的事情，进入语言模型的实际提示是什么？ \n", @@ -653,7 +647,6 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "8dee0f24", "metadata": {}, @@ -669,13 +662,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "7b37c7bc", "metadata": {}, "source": [ - "##### 如何评估新创建的实例\n", - "与创建它们类似，可以运行链条来处理所有示例，然后查看输出并尝试弄清楚，发生了什么，它是否正确" + "**如何评估新创建的实例**: 与创建它们类似，可以运行链条来处理所有示例，然后查看输出并尝试弄清楚，发生了什么，它是否正确" ] }, { @@ -692,12 +683,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "d5bdbdce", "metadata": {}, "source": [ - "## 通过LLM进行评估实例" + "## 6.3 通过LLM进行评估实例" ] }, { @@ -872,12 +862,11 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "9ad64f72", "metadata": {}, "source": [ - "##### 评估思路\n", + "### 6.3.1 评估思路\n", "当它面前有整个文档时，它可以生成一个真实的答案，我们将打印出预测的答，当它进行QA链时，使用embedding和向量数据库进行检索时，将其传递到语言模型中，然后尝试猜测预测的答案，我们还将打印出成绩，这也是语言模型生成的。当它要求评估链评估正在发生的事情时，以及它是否正确或不正确。因此，当我们循环遍历所有这些示例并将它们打印出来时，可以详细了解每个示例" ] }, @@ -950,18 +939,25 @@ ] }, { - "attachments": {}, "cell_type": "markdown", "id": "87ecb476", "metadata": {}, "source": [ - "#### 结果分析\n", + "### 6.3.2 结果分析\n", "对于每个示例，它看起来都是正确的，让我们看看第一个例子。\n", "这里的问题是，舒适的套头衫套装，有侧口袋吗？真正的答案，我们创建了这个，是肯定的。模型预测的答案是舒适的套头衫套装条纹，确实有侧口袋。因此，我们可以理解这是一个正确的答案。它将其评为正确。 \n", - "#### 使用模型评估的优势\n", + "### 6.3.3 使用模型评估的优势\n", "\n", "你有这些答案，它们是任意的字符串。没有单一的真实字符串是最好的可能答案，有许多不同的变体，只要它们具有相同的语义，它们应该被评为相似。如果使用正则进行精准匹配就会丢失语义信息，到目前为止存在的许多评估指标都不够好。目前最有趣和最受欢迎的之一就是使用语言模型进行评估。" ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "e4014482", + "metadata": {}, + "outputs": [], + "source": [] } ], "metadata": { @@ -980,7 +976,25 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.9.16" + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": { + "height": "calc(100% - 180px)", + "left": "10px", + "top": "150px", + "width": "261.818px" + }, + "toc_section_display": true, + "toc_window_display": true } }, "nbformat": 4, diff --git a/content/LangChain for LLM Application Development/7.代理.ipynb b/content/LangChain for LLM Application Development/7.代理.ipynb index b7fe19b..c54fa3a 100644 --- a/content/LangChain for LLM Application Development/7.代理.ipynb +++ b/content/LangChain for LLM Application Development/7.代理.ipynb @@ -5,20 +5,29 @@ "id": "2caa79ba-45e3-437c-9cb6-e433f443f0bf", "metadata": {}, "source": [ - "# 代理\n", + "# 7. 代理\n", + "\n", + "
    \n", + " \n", + "
    \n", + "\n", "\n", "大语言模型学习并记住许多的网络公开信息，大语言模型最常见的应用场景是，将它当作知识库，让它对给定的问题做出回答。\n", "\n", "另一种思路是将大语言模型当作推理引擎，让它基于已有的知识库，并利用新的信息（新的大段文本或者其他信息）来帮助回答问题或者进行推理LongChain的内置代理工具便是适用该场景。\n", "\n", - "本节我们将会了解什么是代理，如何创建代理， 如何使用代理，以及如何与不同类型的工具集成，例如搜索引擎。\n", - "\n", - "**目录**\n", - "* 使用LangChain内置工具\n", - " * 使用llm-math和wikipedia工具\n", - " * 使用PythonREPLTool工具\n", - "* 定义自己的工具并在代理中使用\n", - " * 创建和使用自定义时间工具" + "本节我们将会了解什么是代理，如何创建代理， 如何使用代理，以及如何与不同类型的工具集成，例如搜索引擎。" ] }, { @@ -42,7 +51,7 @@ "id": "631c764b-68fa-483a-80a5-9a322cd1117c", "metadata": {}, "source": [ - "## LangChain内置工具" + "## 7.1 LangChain内置工具" ] }, { @@ -76,7 +85,7 @@ "id": "d5a3655c-4d5e-4a86-8bf4-a11bd1525059", "metadata": {}, "source": [ - "### 📚 使用llm-math和wikipedia工具" + "### 7.1.1 使用llm-math和wikipedia工具" ] }, { @@ -332,7 +341,7 @@ "id": "5901cab6-a7c9-4590-b35d-d41c29e39a39", "metadata": {}, "source": [ - "### 📚 使用PythonREPLTool工具" + "### 7.1.2 使用PythonREPLTool工具" ] }, { @@ -656,7 +665,7 @@ "tags": [] }, "source": [ - "## 定义自己的工具并在代理中使用" + "## 7.2 定义自己的工具并在代理中使用" ] }, { @@ -675,7 +684,7 @@ "id": "fb92f6e4-21ab-494d-9c22-d0678050fd37", "metadata": {}, "source": [ - "### 📚 创建和使用自定义时间工具" + "### 7.2.1 创建和使用自定义时间工具" ] }, { @@ -852,6 +861,19 @@ "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true } }, "nbformat": 4, diff --git a/content/LangChain for LLM Application Development/8.总结.ipynb b/content/LangChain for LLM Application Development/8.总结.ipynb new file mode 100644 index 0000000..1a3e885 --- /dev/null +++ b/content/LangChain for LLM Application Development/8.总结.ipynb @@ -0,0 +1,64 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "87f7cfaa", + "metadata": {}, + "source": [ + "# 8. 总结\n", + "\n", + "\n", + "本次简短课程涵盖了一系列LangChain的应用实践，包括处理顾客评论和基于文档回答问题，以及通过LLM判断何时求助外部工具 (如网站) 来回答复杂问题。\n", + "\n", + "**👍🏻 LangChain如此强大**\n", + "\n", + "构建这类应用曾经需要耗费数周时间，而现在只需要非常少的代码，就可以通过LangChain高效构建所需的应用程序。LangChain已成为开发大模型应用的有力范式，希望大家拥抱这个强大工具，积极探索更多更广泛的应用场景。\n", + "\n", + "**🌈 不同组合, 更多可能性**\n", + "\n", + "LangChain还可以协助我们做什么呢：基于CSV文件回答问题、查询sql数据库、与api交互，有很多例子通过Chain以及不同的提示（Prompts）和输出解析器（output parsers）组合得以实现。\n", + "\n", + "**💪🏻 出发 去探索新世界吧**\n", + "\n", + "因此非常感谢社区中做出贡献的每一个人，无论是协助文档的改进，还是让其他人更容易上手，还是构建新的Chain打开一个全新的世界。\n", + "\n", + "如果你还没有这样做，快去打开电脑，运行 pip install LangChain，然后去使用LangChain、搭建惊艳的应用吧~\n", + "\n" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.12" + }, + "toc": { + "base_numbering": 1, + "nav_menu": {}, + "number_sections": false, + "sideBar": true, + "skip_h1_title": false, + "title_cell": "Table of Contents", + "title_sidebar": "Contents", + "toc_cell": false, + "toc_position": {}, + "toc_section_display": true, + "toc_window_display": true + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/content/LangChain for LLM Application Development/8.课程总结.md b/content/LangChain for LLM Application Development/8.课程总结.md deleted file mode 100644 index ce57191..0000000 --- a/content/LangChain for LLM Application Development/8.课程总结.md +++ /dev/null @@ -1,20 +0,0 @@ -## 吴恩达 LangChain大模型应用开发 总结篇 - -## LangChain for LLM Application Development - -本次简短课程涵盖了一系列LangChain的应用实践，包括处理顾客评论和基于文档回答问题，以及通过LLM判断何时求助外部工具 (如网站) 来回答复杂问题。 - -### 👍🏻 LangChain如此强大 - -构建这类应用曾经需要耗费数周时间，而现在只需要非常少的代码，就可以通过LangChain高效构建所需的应用程序。LangChain已成为开发大模型应用的有力范式，希望大家拥抱这个强大工具，积极探索更多更广泛的应用场景。 - -### 🌈 不同组合->更多可能性 - -LangChain还可以协助我们做什么呢：基于CSV文件回答问题、查询sql数据库、与api交互，有很多例子通过Chain以及不同的提示（Prompts）和输出解析器（output parsers）组合得以实现。 - -### 💪🏻 出发~去探索新世界吧~ - -因此非常感谢社区中做出贡献的每一个人，无论是协助文档的改进，还是让其他人更容易上手，还是构建新的Chain打开一个全新的世界。 - -如果你还没有这样做，快去打开电脑，运行 pip install LangChain，然后去使用LangChain、搭建惊艳的应用吧~ - diff --git a/content/LangChain for LLM Application Development/readme.md b/content/LangChain for LLM Application Development/readme.md index 2a99c32..ea3b506 100644 --- a/content/LangChain for LLM Application Development/readme.md +++ b/content/LangChain for LLM Application Development/readme.md @@ -4,7 +4,7 @@ ### 目录 1. 简介 Introduction @Sarai -2. 模型，提示和解析器 Models, Prompts and Output Parsers @Joye +2. 模型、提示和解析器 Models, Prompts and Output Parsers @Joye 3. 存储 Memory @徐虎 4. 模型链 Chains @徐虎 5. 基于文档的问答 Question and Answer @苟晓攀