|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/generative-ai-and-large-language-models/
课程评论：没有评论
课程名称：生成性人工智能与大型语言模型 概述：本课程为初学者提供了一个实践导向的生成性人工智能和大型语言模型（LLMs）的入门介绍。课程覆盖了机器学习的基础概念以及真实世界的自然语言处理应用，学习者将通过使用Python和Hugging Face获得理论知识与实践经验。到课程结束时，您将理解大型语言模型的工作原理、构建方法，并能够将其应用于如聊天机器人、情感分析和翻译等实际问题。 学习内容： - 机器学习（ML）和生成性人工智能的基础 - 机器学习的定义及现实世界中的应用实例 - 生成与判别性人工智能的区别 - 基本概率概念及贝叶斯定理 - 数字识别的案例研究 - 大型语言模型（LLMs）介绍 - LLM的定义及其功能 - LLM的现实应用 - 理解语言建模的挑战 - LLM背后的核心架构 - 全连接神经网络及其在机器学习中的作用 - 循环神经网络（RNN）及其处理长序列的局限性 - Transformer架构及其优点 - 关键组成部分：分词、嵌入和编码器-解码器模型 - 理解Transformer中的关键概念 - 自注意力机制与QKV矩阵 - Python中的分词和嵌入演示 - 预训练与微调的简单解释 - 推理调整参数：top-k、top-p、温度 - 实践实验与演示 - 实验1：使用Hugging Face构建聊天机器人 - 实验2：对文本数据进行情感分析 - 实验3：创建简单的翻译模型 - 实时Python演示分词、嵌入和推理 - 评估与推理技术 - 模型输出评估的BLEU与ROUGE评分 - 上下文学习：零样本、一样本和少样本的实例 适合人群： - 对AI/ML感兴趣的初学者，寻找生成性语言模型的实用介绍 - 希望了解像ChatGPT这样的模型工作原理的开发者 - 希望以项目为基础学习自然语言处理和生成性人工智能的学生 - 有兴趣使用开源工具构建语言相关应用的任何人 本课程结合了直观的解释、真实的演示和实践实验，确保您在处理大型语言模型和生成性人工智能时具有自信和能力。
This course offers a hands-on, beginner-friendly introduction to Generative AI and Large Language Models (LLMs). From foundational machine learning concepts to real-world NLP applications, learners will gain both theoretical knowledge and practical experience using Python and Hugging Face.By the end of the course, you will understand how LLMs work, how they are built, and how to apply them to real-world problems like chatbots, sentiment analysis, and translation.What You'll Learn:Foundations of Machine Learning (ML) and Generative AIWhat is ML with real-world examplesGenerative vs Discriminative AIBasic probability concepts and Bayes' theoremCase studies in digit recognitionIntroduction to Large Language Models (LLMs)What LLMs are and what they can doReal-world applications of LLMsUnderstanding the language modeling challengeCore Architectures Behind LLMsFully Connected Neural Networks and their role in MLRNNs and their limitations in handling long sequencesTransformer architecture and its advantagesKey components: Tokenization, Embeddings, and Encoder-Decoder modelsUnderstanding Key Concepts in TransformersSelf-Attention mechanism and QKV matricesTokenization and embedding demo in PythonPretraining vs Finetuning explained simplyInference tuning parameters: top-k, top-p, temperatureHands-On Labs and DemosLab 1: Build a chatbot using Hugging FaceLab 2: Perform sentiment analysis on text dataLab 3: Create a simple translation modelLive Python demos on tokenization, embeddings, and inferencingEvaluation and Inference TechniquesBLEU and ROUGE scores for evaluating model outputsIn-context learning: zero-shot, one-shot, and few-shot examplesWho This Course Is For:Beginners in AI/ML looking for a practical introduction to LLMsDevelopers curious about how models like ChatGPT workStudents seeking a project-based approach to NLP and Generative AIAnyone interested in building their own language-based applications using open-source toolsThis course combines intuitive explanations, real-world demos, and hands-on labs to ensure you walk away with both confidence and competence in working with LLMs and Generative AI.
