Illustrated with pictures and texts, 700 pages of machine learning notes are popular! Worth learning

I'm learning machine learning recently , I saw this note , The presentation was very detailed , Record it as a study .

author

Liang Jin （Jim Liang）, come from SAP （ The world's largest commercial software company ）.

Book features

Clarity of organization , It's easier to understand with graphical representation , There are detailed comments on the formula, etc .

Contents summary

Mainly divided into the basic concepts 、 Common algorithms and three other parts .

Why is that? ？

• The first is mathematics , It's about statistics 、 Differential and integral calculus 、 probability 、 Linear algebra, etc , Although everyone has studied advanced mathematics , But if you remember the details , You are a cow . It's more likely that , Most people forget about advanced mathematics , Faced with a large number of formulas in various algorithms , be struck with abhorrenc , Even fear .
• Second, because machine learning itself is a comprehensive discipline , And it is a rapidly developing discipline , The knowledge points are scattered , Lack of systematicness .
• Machine learning on the market / Study books in depth 、 article 、 course , Blossom everywhere , But can express in a clear way 、 A step-by-step tutorial , Not much , A large number of tutorials do not take into account the foundation of learners , Make beginners feel frustrated and confused .
• It is the pain in the process of machine learning that I have personal experience , author Jim Liang Hope to do a tutorial , Explain it in an easy to understand way , Lower the learning threshold for everyone . It took months to do this , Often late at night , I compiled my study notes into this tutorial .

Part 1 Basic concepts are introduced , Include ：

• The process of machine learning
• Data processing
• modeling
• Evaluation indicators （ Such as MSE、ROC curve ）
• Model deployment
• Excessive fitting
• Regularization, etc

In the first part , The author first introduces the machine learning which is widely used nowadays ： From autopilot 、 Voice assistant to robot . Some of these ideas , It is also known by many readers , for example ： Why is machine learning so popular at this time （ big data 、 Computing power 、 Better algorithm ）; machine learning 、 Artificial intelligence 、 The relationship among the three in-depth learning .

In addition to these basic concepts , This tutorial also shows the development process of machine learning model graphically （ Here's the picture ）, Even readers who don't know much about it , You can also learn from this process .

machine learning 700 Electronic version of page notes ：

official account 【 The computer vision Alliance 】 The background to reply ：9001, Electronic version available

stay Part2, The author introduces the commonly used algorithms , Include ：

• Linear regression
• Logical regression
• neural network
• SVM
• Knn
• K-Means
• Decision tree
• Random forests
• AdaBoost
• Naive Bayes
• gradient descent
• Principal component analysis

This part contains a lot of mathematical formulas , But the author tried his best to annotate every formula , Thus sufficient 、 It clearly expresses many mathematical concepts .

For example, in 「 neural network 」 part , The author has arranged 59 Page notes （ from 311 Page to 369 page ）. The author starts with the structure of neurons in the human brain , The artificial neural network is introduced （ANN）、 How artificial neurons work . This note pays great attention to the conceptual explanation of visualization , It is very intuitive to understand .

for example , The concept explanation in the figure below vividly shows the similarity between the working methods of biological neurons and artificial neurons .

Dendritic input of biological neurons - Comparison of axon output mode and input-output mode of artificial neuron .

Over fitting interpretation

When it comes to mathematical formulas , The author will have detailed notes next to it , As shown in the figure below ：

For parallel options （ Such as activation function 、 Common neural network architecture, etc ）, There will also be a comprehensive list ：

For the more complex concepts in neural networks （ Such as seeking guidance 、 Back propagation ）, A few pictures can explain clearly ：

Back propagation algorithm complete process .

Calculation details of forward propagation part .

For your convenience , We have prepared a full version of the machine learning notes PDF, Interested students can follow the following steps to obtain ：

machine learning 700 Electronic version of page notes ：

official account 【 The computer vision Alliance 】 The background to reply ：9001, Electronic version available