Merge ad245b4b38 into 5d57ab9b4e
This commit is contained in:
commit
173019c194
|
|
@ -0,0 +1,35 @@
|
||||||
|
# CSE234: Data Systems for Machine Learning
|
||||||
|
|
||||||
|
## Descriptions
|
||||||
|
|
||||||
|
- Offered by: UCSD
|
||||||
|
- Prerequisites: Linear Algebra, Deep Learning, Operating Systems
|
||||||
|
- Programming Languages: Python, Triton
|
||||||
|
- Difficulty: 🌟🌟🌟
|
||||||
|
- Class Hour: 80 hours
|
||||||
|
|
||||||
|
<!--
|
||||||
|
Introduce the course in a paragraph or two, including but not limited to:
|
||||||
|
(1) The technical knowledge covered in lectures
|
||||||
|
(2) Its differences and features compared to similar courses
|
||||||
|
(3) Your personal experiences and feelings after studying this course
|
||||||
|
(4) Caveats about studying this course on your own (pitfalls, difficulty warnings, etc.)
|
||||||
|
(5) ... ...
|
||||||
|
-->
|
||||||
|
|
||||||
|
|
||||||
|
This course is focused on designing a wholistic LLM System class as an introduction to design efficient systems for LLM.
|
||||||
|
|
||||||
|
The class into three parts, covering the following topics.
|
||||||
|
|
||||||
|
1. Basics: deep learning, autodiff, CUDA programming, ML hardware
|
||||||
|
2. ML systems and optimizations: Dataflow graph systems, ML compilation, memory and graph optimization, ML parallelism, auto-parallelization
|
||||||
|
3. LLM systems: LLM training, data curation, inference and serving, attention optimization, scaling law, RAG, LLM agents
|
||||||
|
|
||||||
|
|
||||||
|
## Course Resources
|
||||||
|
|
||||||
|
- Course Website: https://hao-ai-lab.github.io/cse234-w25/
|
||||||
|
- Recordings: https://hao-ai-lab.github.io/cse234-w25/
|
||||||
|
- Textbooks: https://hao-ai-lab.github.io/cse234-w25/resources/
|
||||||
|
- Assignments: https://hao-ai-lab.github.io/cse234-w25/assignments/
|
||||||
|
|
@ -0,0 +1,35 @@
|
||||||
|
# CSE234: Data Systems for Machine Learning
|
||||||
|
|
||||||
|
|
||||||
|
## 课程简介
|
||||||
|
|
||||||
|
- 所属大学:UCSD
|
||||||
|
- 先修要求:线性代数,深度学习,操作系统
|
||||||
|
- 编程语言:Python, Triton
|
||||||
|
- 课程难度:🌟🌟🌟
|
||||||
|
- 预计学时:80小时
|
||||||
|
|
||||||
|
<!-- 用一两段话介绍这门课程,内容包括但不限于:
|
||||||
|
(1)课程覆盖的知识点范围
|
||||||
|
(2)与同类课程相比它的优势与特点
|
||||||
|
(3)学习这门课程的体验与感受
|
||||||
|
(4)自学这门课的注意点(踩过的坑、难度预警等等)
|
||||||
|
(5)... ...
|
||||||
|
-->
|
||||||
|
|
||||||
|
本课程专注于设计一个全面的大语言模型(LLM)系统课程,作为设计高效LLM系统的入门介绍。
|
||||||
|
|
||||||
|
课程分为三个部分,涵盖以下主题:
|
||||||
|
|
||||||
|
1. 基础知识:深度学习、自动微分、CUDA编程、机器学习硬件
|
||||||
|
2. 机器学习系统与优化:数据流图系统、机器学习编译、内存与图优化、机器学习并行化、自动并行化
|
||||||
|
3. 大语言模型系统:LLM训练、数据整理、推理与服务、注意力机制优化、缩放定律、检索增强生成(RAG)、Agent
|
||||||
|
|
||||||
|
|
||||||
|
## 课程资源
|
||||||
|
|
||||||
|
- 课程网站:https://hao-ai-lab.github.io/cse234-w25/
|
||||||
|
- 课程视频:https://hao-ai-lab.github.io/cse234-w25/
|
||||||
|
- 课程教材:https://hao-ai-lab.github.io/cse234-w25/resources/
|
||||||
|
- 课程作业:https://hao-ai-lab.github.io/cse234-w25/assignments/
|
||||||
|
|
||||||
Loading…
Reference in New Issue