|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/llm-rag-langchain-llamaindex/
课程评论：没有评论
**课程名称（暂译）：** 生成式AI - LLM RAG 两大框架整合（LangChain + LlamaIndex） **课程概述：** 本课程深入探讨如何结合利用LangChain和LlamaIndex两大主流框架，并整合OpenAI GPT、Google Gemini API以及ChromaDB、Pinecone等向量数据库，来构建高级的LLM RAG（检索增强生成）应用。课程内容包含详尽的概念解析和实践操作，旨在帮助学员全面掌握LLM RAG应用的核心要素。 课程将清晰、简洁地剖析两大框架的关键组件，包括Agents、Tools、Chains、Memory、QueryPipelines、Retrievers和Query Engines。学员还将深入学习语言嵌入（Language Embeddings）和向量数据库，从而能够开发高效的语义搜索和基于相似度的RAG应用。此外，课程还涵盖了多种Prompt Engineering技术，以提升RAG应用的性能。 **项目/实践环节（部分）：** * **会话式记忆聊天机器人：** 使用下载的网页数据和向量数据库构建。 * **简历上传与语义搜索应用：** 实现简历内容的智能搜索。 * **发票提取RAG应用：** 针对发票信息进行提取。 * **结构化数据分析应用：** 支持自然语言查询数据分析。 * **ReAct Agent：** 使用ReAct Agent和Tools创建计算器应用。 * **文档Agent与动态工具：** 动态创建QueryEngineTools，并通过Agent进行查询编排。 * **顺序查询Pipeline：** 构建简单的顺序查询管线。 * **DAG Pipeline：** 开发复杂的有向无环图（DAG）管线。 * **Dataframe Pipeline：** 利用Pandas、Output Parser和Response Synthesizer开发复杂的数据框分析管线。 * **SQL数据库互动：** 构建SQL数据库的摄取与查询机器人。 **课程亮点：** 本课程采用的双框架方法，能为学员提供更广阔的RAG开发视角，使其能够充分发挥LangChain和LlamaIndex各自的优势。
This course leverages the power of both LangChain and LlamaIndex frameworks, along with OpenAI GPT and Google Gemini APIs, and Vector Databases like ChromaDB and Pinecone. It is designed to provide you with a comprehensive understanding of building advanced LLM RAG applications through in-depth conceptual learning and hands-on sessions. The course covers essential aspects of LLM RAG apps, exploring components from both frameworks such as Agents, Tools, Chains, Memory, QueryPipelines, Retrievers, and Query Engines in a clear and concise manner. You'll also delve into Language Embeddings and Vector Databases, enabling you to develop efficient semantic search and similarity-based RAG applications. Additionally, the course covers various Prompt Engineering techniques to enhance the efficiency of your RAG applications.List of Projects/Hands-on included: Develop a Conversational Memory Chatbot using downloaded web data and Vector DBCreate a CV Upload and Semantic CV Search App Invoice Extraction RAG AppCreate a Structured Data Analytics App that uses Natural Language Queries ReAct Agent: Create a Calculator App using a ReAct Agent and ToolsDocument Agent with Dynamic Tools: Create multiple QueryEngineTools dynamically and orchestrate queries through AgentsSequential Query Pipeline: Create Simple Sequential Query PipelinesDAG Pipeline: Develop complex DAG PipelinesDataframe Pipeline: Develop complex Dataframe Analysis Pipelines with Pandas Output Parser and Response SynthesizerWorking with SQL Databases: Develop SQL Database ingestion BotThis twin-framework approach will provide you with a broader perspective on RAG development, allowing you to leverage the strengths of both LangChain and LlamaIndex in your projects.
