Grokking Deep Learning teaches you to build deep learning neural networks from scratch! In his engaging style, seasoned deep learning expert Andrew Trask shows you the science under the hood, so you grok for yourself every detail of training neural networks. Using only Python and its math-supporting library, NumPy, you'll train your own neural networks to see and understand ima...
Grokking Deep Learning teaches you to build deep learning neural networks from scratch! In his engaging style, seasoned deep learning expert Andrew Trask shows you the science under the hood, so you grok for yourself every detail of training neural networks. Using only Python and its math-supporting library, NumPy, you'll train your own neural networks to see and understand images, translate text into different languages, and even write like Shakespeare! When you're done, you'll be fully prepared to move on to mastering deep learning frameworks.
作者简介
· · · · · ·
Andrew Trask is a PhD student at Oxford University and a research scientist at DeepMind. Previously, Andrew was a researcher and analytics product manager at Digital Reasoning, where he trained the world's largest artificial neural network and helped guide the analytics roadmap for the Synthesys cognitive computing platform.
目录
· · · · · ·
Introducing deep learning: why you should learn it
Fundamental concepts: how do machines learn?
Introduction to neural prediction: forward propagation
Introduction to neural learning: gradient descent
Learning multiple weights at a time: generalizing gradient descent
Building your first deep neural network: introduction to backpropagation
· · · · · ·
(更多)
Introducing deep learning: why you should learn it
Fundamental concepts: how do machines learn?
Introduction to neural prediction: forward propagation
Introduction to neural learning: gradient descent
Learning multiple weights at a time: generalizing gradient descent
Building your first deep neural network: introduction to backpropagation
How to picture neural networks: in your head and on paper
Learning signal and ignoring noise:introduction to regularization and batching
Modeling probabilities and nonlinearities: activation functions
Neural learning about edges and corners: intro to convolutional neural networks
Neural networks that understand language: king - man + woman == ?
Neural networks that write like Shakespeare: recurrent layers for variable-length data
Introducing automatic optimization: let's build a deep learning framework
Learning to write like Shakespeare: long short-term memory
Deep learning on unseen data: introducing federated learning
Where to go from here: a brief guide
· · · · · · (收起)
原文摘录
· · · · · ·
for i in range(len(true)):
error[i] = (pred[i] - true[i]) ** 2
delta = pred[i] - true[i] (查看原文)
The structure trick
When a neural network needs to use the same idea in multiple places, endeavor to use the same weights in both places. This will make those weights more intelligent by giving them more samples to learn from, increasing generalization. (查看原文)
0 有用 永远自由的心 2020-07-18 14:05:43
适合入门
9 有用 泥豆尼痘昵 2019-04-15 20:58:13
首先，这书是19年出版，不是17年；第二，这本书只有前半部分适合入门，中间往后越来越有偷懒之嫌，我读过较多深度学习入门书，无论是偏理论还是偏实践，这是第一本介绍LSTM不用一张图，而是完全用代码堆砌和打不到重点的文字，LSTM明明最容易且最有必要用图释呈现，这种情况在书中的后半部分很常见，不止是针对LSTM，作者一改在前半部分中动不动就作图说明的风格，那时候甚至不太重要的概念都作图，后来是感觉读者... 首先，这书是19年出版，不是17年；第二，这本书只有前半部分适合入门，中间往后越来越有偷懒之嫌，我读过较多深度学习入门书，无论是偏理论还是偏实践，这是第一本介绍LSTM不用一张图，而是完全用代码堆砌和打不到重点的文字，LSTM明明最容易且最有必要用图释呈现，这种情况在书中的后半部分很常见，不止是针对LSTM，作者一改在前半部分中动不动就作图说明的风格，那时候甚至不太重要的概念都作图，后来是感觉读者们翅膀一下子就硬了，不需要深入浅出了？可你也没什么过渡部分啊，代码数量在书的后半部分突然猛增，最重要的是注释写的真烂，号称是入门书，代码注释这么烂你好意思吗？能简简单单看懂你代码的人还需要读入门书吗？给三星评价是因为书的前半段有可取之处，在大量同类书中难得用代码把一些最基础功能实现，而不是外调函数。 (展开)
1 有用 swq457080 2020-08-27 09:26:19
Andrew Ng的Deep learning 替代本
0 有用 豆友199285186 2021-10-20 12:24:32
挺好的书，前十章是基础，从第十一章开始，难度陡然增加。我建议所有的初学者都应该把里面的代码都学会了，自己学习写个框架，然后自己操作，再去学习类似于pytorch，keras等框架。
0 有用 一只黑眼睛看着大千世界 2021-06-03 23:28:41
示意图+基础代码简单入门