【导读】斯坦福大学在2017年开设了一门深度学习Tensorflow实战课程(Tensorflow for Deep Learning Research),广受好评,2018年课程最近已经开始,课程都会提供丰富的学习资源,比如PPT, 讲义,代码和课程报告,是非常好的Tensorflow实战入门课程。该课程没有以传统的稳扎稳打的学习方式(先学习数学基础、然后掌握机器学习、最后开始深度学习),而是直接从当前最热门的深度学习框架Tensorflow入手,讲解如何利用开源的Tensorflow框架进行深度学习研究,促使读者在实战中学习和掌握深度学习.因此,本教程更适合已经有一定深度学习基础、希望快速构建项目的读者。
▌课程描述
TensorFlow是由Google的研究人员开发的一个功能强大的机器学习开源工具库。它功能强大,可以简化构建不同神经网络的任务。 TensorFlow允许在不同计算机之间分配计算,并在一台机器中分配多个CPU和GPU,以实现并行。 TensorFlow提供了一个Python API,以及一个C ++ API(C++ API文档较少)。在本课程中,我们将使用Python API。
本课程涵盖TensorFlow库的基本使用,以及使用TensorFlow库进行深度学习研究。我们旨在帮助学生理解TensorFlow的计算模型,探索TensorFlow提供的功能,并学习如何构建最适合的模型。通过深度学习项目。通过此课程,学生将使用TensorFlow建立不同复杂度的模型,从简单的线性/逻辑回归到卷积神经网络和递归神经网络,以解决词嵌入,翻译,光学字符识别,强化学习等任务,学生还可以学习如何构建模型。
Event | Date | Description | ||
---|---|---|---|---|
Jan 10 Week 1 |
No class | |||
Lecture | Jan 12 | Overview of Tensorflow Why Tensorflow? Graphs and Sessions |
||
To do | Jan 12 | Check out TensorBoard | ||
Lecture | Jan 17 Week 2 |
Operations Basic operations, constants, variables Control dependencies Data pipeline TensorBoard |
||
Workshop | Jan 19 | Linear and Logistic Regression Tensorflow's Optimizers Example: OCR task on MNIST dataset |
||
A1 released | Jan 19 | Assignment #1 released | ||
Lecture | Jan 24 Week 3 |
Structure your TensorFlow model Example: word2vec |
||
Lecture | Jan 26 | Manage experiments and process data Interfaces Saver object, checkpoints Example: word2vec |
||
A1 Due | Jan 30 | Assignment #1 due | ||
Lecture | Jan 31 Week 4 |
Convolutional Neural Networks Example: Image Classification |
||
Lecture | Feb 2 | Convolutional Neural Networks Example: Autoencoder |
||
A2 released | Feb 7 | Assignment #2 released | ||
Lecture | Feb 7 Week 5 |
Convolutional Neural Networks Discussion of Assignment #2 Example: Style Transfer |
||
Lecture | Feb 9 | Debug your TensorFlow model | ||
Feb 14 Week 6 |
GANs Guest lecture by Alec Radford (Research Scientist at OpenAI) |
|||
Lecture | Feb 16 | Midterm discussion | ||
A2 Due | Feb 19 | Assignment #2 due | ||
Lecture | Feb 20 Week 7 |
Recurrent Neural Networks Example: Character-level Language Modeling |
||
Lecture | Feb 23 | Seq2seq with Attention Example: Neural machine translation |
||
Lecture | Feb 28 Week 8 |
Beyond RNNs | ||
A3 released | Mar 1 | Assignment #3 released | ||
Mar 2 | Open-domain dialogue system Training and Optimizing |
|||
Lecture | Mar 7 Week 9 |
Reinforcement Learning in Tensorflow | ||
Lecture | Mar 9 | Keras Guest lecture by François Chollet (Deep learning researcher at Google, author of Keras) |
||
A3 Due | Mar 16 | Assignment #3 due | ||
Demo | Mar 16 | Demo |
参考链接:
1. 新版CS 20:Tensorflow for Deep Learning Research
https://web.stanford.edu/class/cs20si/index.html
2. 旧版CS 20SI: Tensorflow for Deep Learning Research
https://web.stanford.edu/class/cs20si/2017/
3. https://web.stanford.edu/class/cs20si/syllabus.html
请关注专知公众号(扫一扫最下面专知二维码,或者点击上方蓝色专知),
后台回复“STDL” 就可以获取 第一次课PPT和Note下载链接~
▌课程介绍(第一次课PPT)
附上第一次课的PPT内容
-END-
专 · 知
人工智能领域主题知识资料查看获取:【专知荟萃】人工智能领域26个主题知识资料全集(入门/进阶/论文/综述/视频/专家等)
同时欢迎各位用户进行专知投稿,详情请点击:
【诚邀】专知诚挚邀请各位专业者加入AI创作者计划!了解使用专知!
请PC登录www.zhuanzhi.ai或者点击阅读原文,注册登录专知,获取更多AI知识资料!
请扫一扫如下二维码关注我们的公众号,获取人工智能的专业知识!
请加专知小助手微信(Rancho_Fang),加入专知主题人工智能群交流!
点击“阅读原文”,使用专知!