AI By Doing

AI By Doing: A Hands-on Guide to Artificial Intelligence

huhuhang

cover

AI By Doing: A Hands-on Guide to Artificial Intelligence#

Preface#

I began my journey into machine learning in 2015, starting with Professor Li Hang’s “Statistical Learning Methods.” As I progressed through various learning materials, I noticed some common challenges in the field.

For instance, many textbooks written by university professors tend to be heavily theoretical and require advanced mathematical knowledge, making them difficult for beginners to grasp. They often lack practical code implementations. On the other hand, books written by engineers typically focus on practical applications, merely introducing tools and libraries without delving into the underlying principles. Additionally, many resources provide code examples alongside algorithmic concepts, but these examples are often incomplete or disconnected from the theory, making it difficult for readers to replicate the results.

When starting with machine learning, simply knowing how to use libraries without understanding the principles can be limiting - you might not even understand parameter settings, let alone know how to tune them. Conversely, if you only know the theory without practical experience, you might struggle to implement even the simplest models. My goal is to help you both understand the principles and apply them in practice, giving you a comprehensive grasp of machine learning.

Since 2018, I’ve been creating content using Jupyter Notebook, which allows for seamless integration of text and code, making it easier to read and understand. It also displays code alongside its output, facilitating practice and replication.

This hands-on AI tutorial series aims to introduce you to machine learning and deep learning, starting with supervised learning. I break down and derive the principles behind each fundamental algorithm, showing the mathematical processes while implementing them in Python code, ensuring a clear correspondence between formulas and implementation. I also demonstrate how to achieve the same results using mainstream frameworks, helping readers understand the differences between manual implementation and framework-based approaches.

This content assumes you have basic Python programming skills and undergraduate-level mathematics knowledge in calculus, linear algebra, and probability theory. Don’t worry if you’re rusty - we’ll help you refresh these concepts as needed. I’ve included “Appendix I: Mathematical Foundations for Machine Learning” and “Appendix II: Common Tools in Machine Learning” for reference. These are in the appendix because I don’t recommend studying them first, as this might consume too much time and diminish your enthusiasm. Instead, refer to them as needed during your learning journey.

The best way to learn from this course is to practice hands-on with your computer. I recommend using Jupyter Notebook to document your code exercises and understanding, creating your own study notes. Simply reading through the material is unlikely to be effective.

Currently, this work comprises 97 chapters (24,150 words) and is continuously updated. In this era of AI-generated content, I hope this human-crafted material helps you better understand machine learning.

I recommend learning basic Linux skills through LabEx in your spare time, as this will greatly help with the practical aspects of this tutorial.

Membership#

If you find this tutorial helpful, consider supporting the Pro membership plan. This is entirely optional - all content remains free to read and learn from.

Pro Plan

Pro Membership Plan

The Pro plan is a membership program launched to maintain and update this tutorial. You’ll receive additional benefits and services - click for details.

Additionally, I’m an indie iOS developer. Feel free to check out and support my Apps on the App Store.

License#

This work is created by huhuhang and is registered and notarized for copyright protection.

This site is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 International license.

Additional rules include:

  • 🚫 Do not clone the entire site to create similar websites

  • 🚫 Do not compile site content into e-books for distribution

  • 🚫 Please don’t “open source” the site content on my behalf

  • 🚫 Do not use the content for commercial purposes (for commercial licensing, contact huhuhang@gmail.com)

The articles contain numerous formulas, code examples, and custom styles. I created this site to provide the best possible learning experience. Rather than copying or reposting content, please share the original link so readers can access the latest version and enjoy the optimal reading experience.

Table of Contents#

Supervised Learning (Regression)#

1. Overview and Examples of Machine Learning

A comprehensive introduction to the basic concepts and classifications of machine learning, exploring supervised and unsupervised learning in depth, and analyzing the relationship between machine learning, deep learning, and artificial intelligence, providing beginners with a comprehensive and systematic machine learning knowledge system.

55 words Last Updated: 2024-11-10

2. Linear Regression Implementation and Application

Details the basic principles, implementation process, and application cases of linear regression. By explaining knowledge points such as univariate linear regression and squared loss functions, it helps readers master the core ideas and practical skills of linear regression.

14 words Last Updated: 2025-02-18

4. Beijing Housing Price Prediction

This tutorial will introduce how to use a linear regression model to predict housing prices in Beijing. Includes dataset reading and partitioning, model training and prediction, and model evaluation steps. Detailed practical guidance and code examples are provided for beginners.

87 words Last Updated: 2025-02-18

5. Polynomial Regression Implementation and Application

This article details the basic principles and implementation methods of polynomial regression, and demonstrates how to use polynomial regression models for nonlinear data fitting and prediction through example data, providing practical technical guidance for machine learning and data analysis practice.

534 words Last Updated: 2025-02-18

6. Bitcoin Price Prediction and Plotting

This course will teach you how to use polynomial regression and ridge regression methods to predict the trend of Bitcoin price changes, and analyze key indicators such as Bitcoin market price, total amount, and transaction fees through visualization. Suitable for machine learning enthusiasts interested in digital currencies.

152 words Last Updated: 2025-02-18

7. Ridge Regression and LASSO Regression Implementation

This course details the principles and implementation of ridge regression and LASSO regression. Through example code, it explains how to apply these two regularization regression methods to address the limitations of linear regression, providing practical guidance for learning machine learning algorithm implementation.

185 words Last Updated: 2025-02-18

8. Calculate Ridge Regression Coefficients Using Matrices

Learn to implement ridge regression using Python matrix operations. Compare results with scikit-learn implementations for comprehensive ML understanding.

79 words Last Updated: 2025-02-18

9. Regression Model Evaluation and Testing

This article details the relevant knowledge points of regression model evaluation and validation, including goodness-of-fit tests, variable significance tests, and common evaluation metrics such as MAE, MSE, and MAPE, and demonstrates with examples how to calculate the goodness-of-fit of models. For readers who want to learn in-depth about machine learning regression models, this is an essential and valuable piece of content.

114 words Last Updated: 2025-02-18

10. Comprehensive Application Exercises of Regression Methods

This machine learning tutorial will combine the relevant regression analysis methods learned earlier to complete a multiple regression analysis task and review the implementation and application of different methods. Includes knowledge points such as univariate linear regression, multiple linear regression, and hypothesis testing.

68 words Last Updated: 2025-02-18

Supervised Learning (Classification)#

11. Logistic Regression Implementation and Application

This course delves into the principles and algorithm implementation of logistic regression, and uses scikit-learn to build a logistic regression classification prediction model, helping learners master the basic classification methods in machine learning.

202 words Last Updated: 2025-02-18

12. Gradient Descent Method Implementation and Application

This article details the application of the gradient descent method in machine learning, including the implementation and optimization of linear regression problems. Through code examples, it helps readers deeply understand the principles and usage of the gradient descent algorithm.

83 words Last Updated: 2025-02-18

13. K-Nearest Neighbors Algorithm Implementation and Application

This article details the principles and implementation of the K-Nearest Neighbors algorithm, explores the KNN algorithm in depth through distance calculation and classification decisions, and applies it to the iris classification task, providing practical learning resources for machine learning beginners.

383 words Last Updated: 2025-02-18

14. Implementation and Application of K-Nearest Neighbor Regression Algorithm

This course details the implementation and application of the K-Nearest Neighbors regression algorithm, including key steps such as data preparation, distance calculation, neighbor searching, and prediction value calculation, helping readers deeply master the practical application of core machine learning algorithms.

66 words Last Updated: 2025-02-18

15. Naive Bayes Implementation and Application

Details the principles and implementation of the Naive Bayes algorithm, including basic knowledge such as Bayes' theorem, conditional probability, prior probability, and posterior probability, and provides a Python code implementation, suitable for beginners to learn and apply.

2195 words Last Updated: 2025-02-18

16. Gaussian Distribution Function Implementation and Plotting

This course will teach you how to implement a Gaussian distribution function using Python and draw Gaussian distribution images under different parameters using Matplotlib. Master the application of Gaussian distributions in machine learning and data analysis.

46 words Last Updated: 2025-02-18

17. Classification Model Evaluation Methods

This article details the commonly used evaluation metrics for classification models, including accuracy, precision, recall, F1-score, and ROC curves. It comprehensively explains how to use these metrics to evaluate the performance of classification models using a credit card risk prediction model as an example. The content is practical and easy to understand, suitable for machine learning beginners.

159 words Last Updated: 2025-02-18

18. Support Vector Machine Implementation and Application

Details the principles and applications of support vector machines, including linear classification support vector machines, Lagrange duality, and kernel tricks, and provides practical code demonstrations. Suitable for beginners and advanced learners, helping you master this powerful machine learning classification algorithm.

99 words Last Updated: 2025-02-18

19. Support Vector Machine for Portrait Classification

This course will use the support vector machine method provided by scikit-learn to complete the human face image classification task. By preprocessing data, extracting features, and training models, you'll learn how to implement human portrait classification using the support vector machine algorithm. The course content is practical and challenging, suitable for learners interested in machine learning.

51 words Last Updated: 2025-02-18

20. Decision Tree Implementation and Application

Master decision tree algorithms through practical implementation. Learn feature selection, tree construction, and visualization for student grade prediction.

534 words Last Updated: 2025-02-18

21. Decision Tree Model Parameter Optimization and Selection

This article details the methods for optimizing and selecting decision tree model parameters, including grid search and random search, and demonstrates with practical cases how to use the scikit-learn library for decision tree model parameter tuning, helping readers master the skills of machine learning model parameter tuning.

81 words Last Updated: 2025-02-18

22. Bagging and Boosting Ensemble Learning Methods

Explore ensemble learning concepts, including Bagging and Boosting algorithms. Learn to improve classification accuracy through practical AI implementations.

140 words Last Updated: 2025-02-18

23. Application of Heterogeneous Integration Voting Method

This article introduces the application of heterogeneous ensemble voting methods in machine learning, including combinations of algorithms such as logistic regression, decision trees, and naive Bayes, to improve classification accuracy through voting. The content is practical and easy to understand, suitable for beginners to learn.

11 words Last Updated: 2025-02-18

24. Quickly Select Models Using Cross-Validation

This course will teach you how to use K-fold cross-validation to quickly select the most suitable machine learning model. Through practical exercises, you will master the principles and applications of cross-validation, and find the best model for your machine learning project.

136 words Last Updated: 2025-02-18

Unsupervised Learning (Clustering)#

25. Implementation and Application of Partitioning Clustering Methods

This course deeply explains partitioning clustering methods, focusing on the representative algorithm K-Means and its variants. Through vivid and visual demonstrations, it helps learners quickly master the principles and implementation of clustering algorithms.

557 words Last Updated: 2025-02-18

26. Image Compression Using K-Means

This article introduces how to use the Mini Batch K-Means algorithm to perform image compression processing, by clustering similar pixels to achieve effective image compression, suitable for learners in the fields of machine learning and image processing.

164 words Last Updated: 2025-02-18

27. Implementation and Application of Hierarchical Clustering Method

This article details the concepts of hierarchical clustering methods, the bottom-up and top-down clustering algorithms, and different distance calculation methods, providing readers with a practical guide for a comprehensive understanding of hierarchical clustering. By studying this article, readers can master the principles and application scenarios of hierarchical clustering, providing a powerful tool for data analysis.

1242 words Last Updated: 2025-02-18

28. Principal Component Analysis Principle and Applications

This article details the principles and applications of Principal Component Analysis (PCA). Starting from the basic knowledge of vector spaces, basis vectors, and matrix transformations, it gradually explains the core ideas and calculation processes of PCA, providing readers with a comprehensive and in-depth PCA learning guide.

4412 words Last Updated: 2025-02-18

29. Hierarchical Clustering Application and Dendrogram Plotting

This article introduces how to use SciPy to perform hierarchical clustering analysis on wheat seed data and draw dendrogram visualization. The content covers the application of hierarchical clustering algorithms and the pruning and visualization of clustering trees, providing practical case analysis for machine learning practice.

33 words Last Updated: 2025-02-18

30. Implementation and Application of Density Clustering Method

Details the principles and implementation of the density clustering algorithm DBSCAN, helping readers master effective methods for non-convex data clustering, and demonstrating its advantages through example applications. Suitable for machine learning beginners and advanced learners.

285 words Last Updated: 2025-02-18

31. Density Clustering to Mark Abnormal Shared Bicycles

Learn to detect anomalies in bike-sharing GPS data using density-based clustering algorithms. Covers DBSCAN and HDBSCAN implementation for real-world applications.

39 words Last Updated: 2025-02-18

32. Application of Spectral Clustering and Other Clustering Methods

This article introduces the principles of spectral clustering algorithms and other common clustering methods, including core concepts such as Laplacian matrices and undirected graph cuts, providing readers with a detailed machine learning tutorial to comprehensively master clustering algorithms.

91 words Last Updated: 2025-02-18

33. Comparative Evaluation of Common Clustering Algorithms

This article details the clustering effects and performance comparison of 8 common clustering algorithms on datasets of different shapes, providing readers with a comprehensive practical experience of machine learning clustering algorithms.

47 words Last Updated: 2025-02-18

Unsupervised Learning (Association Rules)#

34. Apriori Association Rule Learning Method

This article introduces the basic concepts and methods of association rule learning, including core knowledge such as frequent itemsets, support, and confidence, and provides practical examples of market basket analysis to help readers deeply understand the application scenarios and implementation steps of association rule learning.

350 words Last Updated: 2025-02-18

35. Association Rule Analysis of Shopping Data

This tutorial will use the Apriori algorithm to analyze association rules for supermarket shopping data, find frequent itemsets and association rules, and help merchants understand customer purchasing habits and optimize product display and promotion strategies.

18 words Last Updated: 2025-02-18

36. Time Series Data Analysis and Processing

Learn to preprocess and analyze time series data using Pandas, including time generation, time conversion, time calculation, time zone management, and other techniques, helping you master the key knowledge of time series data analysis.

246 words Last Updated: 2025-02-18

37. Stock Time Series Data Processing

This challenge will use stock time series data to build a model through data processing and feature engineering, predicting the rise and fall of stocks. Learn knowledge points such as time format conversion, time series data resampling, and data sorting.

11 words Last Updated: 2025-02-18

38. Time Series Data Modeling and Analysis

This article details the characteristics and classification of time series data, and explores key concepts and modeling methods such as descriptive time series analysis, statistical time series analysis, and stationary time series tests, providing readers with a comprehensive practical guide to time series data analysis and modeling.

146 words Last Updated: 2025-02-18

39. Agricultural Production Index Modeling Analysis

This challenge will attempt to use machine learning models to model and analyze the agricultural production index, and predict future agricultural production indices using historical data. Includes data preprocessing, data resampling, and the use of Prophet knowledge points.

4 words Last Updated: 2025-02-18

Machine Learning Engineering (Model Deployment and Inference)#

40. A Review of Automated Machine Learning

Automated machine learning is a popular branch of machine learning that can significantly reduce the development threshold of machine learning models. This article summarizes the concepts, goals, and key technologies of automated machine learning, comprehensively introducing this artificial intelligence solution.

0 words Last Updated: 2025-02-18

41. Practical Applications of Automated Machine Learning

This article introduces auto-sklearn, an automated machine learning framework based on scikit-learn, including classification algorithms, regression algorithms, and the advantages and disadvantages of automated machine learning, providing a convenient machine learning modeling method for non-professional developers.

63 words Last Updated: 2025-02-18

42. Handwritten Character Classification with AutoML

This tutorial will take you through the MNIST handwritten character classification task using the auto-sklearn framework. Through Kaggle competition practice, you will master the application skills of automated machine learning and improve the efficiency of machine learning modeling.

0 words Last Updated: 2025-02-18

43. Machine Learning Model Inference and Deployment

This article details the inference and deployment process of machine learning models, including model saving, model deployment, and dynamic inference, providing practical operation guides for machine learning engineers.

123 words Last Updated: 2025-02-18

44. Mushroom Classification Model Deployment and Inference

This tutorial will help you understand how to train a poisonous mushroom classification model using scikit-learn and deploy it as a Web application using Flask. You will learn the entire process of model saving, deployment, and inference, providing a reference for your machine learning project practice.

203 words Last Updated: 2025-02-18

45. Dynamic Incremental Training of Machine Learning Models

Learn how to use artificial neural networks to implement dynamic incremental training of machine learning models and deploy them online for real-time inference, applicable to scenarios such as handwritten character recognition. Master key technologies such as dynamic model training and incremental learning.

520 words Last Updated: 2025-02-18

46. Online Learning and Cloud Model Deployment

This challenge focuses on online incremental learning of models. You need to complete the online deployment of the model as required and enable it to infer and incrementally learn from new data. Learn machine learning practical skills such as online learning and cloud deployment.

99 words Last Updated: 2025-02-18

Deep Learning Principles (Artificial Neural Networks)#

47. Overview and Examples of Deep Learning

This article deeply introduces the concepts and characteristics of deep learning, and provides a detailed analysis of the relationship between machine learning, deep learning, and artificial intelligence, helping readers comprehensively master the development history and application prospects of deep learning.

0 words Last Updated: 2025-02-18

48. Perceptron and Artificial Neural Network

Explore the origins, characteristics, and applications of perceptrons and artificial neural networks, learn the basic knowledge of the machine learning field, and lay a solid foundation for deep learning.

364 words Last Updated: 2025-02-18

49. Handwritten Character Recognition Neural Network

This article details the principles and implementation of a handwritten character recognition neural network, covering core knowledge points such as artificial neural networks and handwritten character recognition, and provides practical code examples to help readers master the artificial intelligence technology of handwritten character recognition.

77 words Last Updated: 2025-02-18

Deep Learning Frameworks (TensorFlow & PyTorch)#

50. Basic Concepts and Syntax of TensorFlow

Learn the fundamental concepts and syntax of TensorFlow, master the key skills for building deep learning models, and assist in the development of artificial intelligence projects. Core knowledge points include tensors, Eager Execution, and an overview of APIs.

81 words Last Updated: 2025-02-18

51. TensorFlow California Housing Price Prediction

This article details how to use the TensorFlow framework to build and train linear regression models. By using a practical case to predict California housing prices, it provides beginners with a practical guide to machine learning.

42 words Last Updated: 2025-02-18

52. Building Neural Networks with TensorFlow

Learn how to build neural networks using TensorFlow, from basic NumPy implementations to DIGITS classification and Mini Batch training. Master core TensorFlow concepts.

619 words Last Updated: 2025-02-18

53. TensorFlow Automotive Assessment Classification

Hands-on practice AI By Doing provides a TensorFlow automotive rating classification tutorial, including data preprocessing, neural network construction, training, and evaluation, helping readers master the application of TensorFlow in the field of machine learning.

29 words Last Updated: 2025-02-18

54. Using TensorFlow High-Level APIs

Comprehensive guide to TensorFlow's advanced APIs, including Keras Sequential models, Functional API, model persistence, and Estimator API usage.

311 words Last Updated: 2025-02-18

55. TensorFlow Fashion Item Classification

Build a DNN network with TensorFlow Keras and complete the fashion item classification task, including knowledge points such as grayscale data standardization, Flatten, and Dropout, bringing you an exciting experience of hands-on artificial intelligence.

10 words Last Updated: 2025-02-18

56. Basic Concepts and Syntax of PyTorch

PyTorch is a powerful and easy-to-use deep learning framework. This course will comprehensively introduce the basic concepts and syntax of PyTorch, including tensor types, index slicing, automatic differentiation, and other core functions, helping you quickly get started with PyTorch and build deep neural networks.

81 words Last Updated: 2025-02-18

57. Building Neural Networks with PyTorch

PyTorch is a powerful deep learning framework, and this tutorial will comprehensively introduce how to build neural network models using PyTorch, including the Sequential container structure, GPU-accelerated training, and model saving and inference.

344 words Last Updated: 2025-02-18

58. PyTorch Implementation of Linear Regression

This article details how to implement a linear regression model using PyTorch, including model building, loss function definition, optimizer settings, and iterative optimization processes, providing readers with valuable experience in practicing machine learning.

70 words Last Updated: 2025-02-18

Deep Learning Applications (Computer Vision)#

59. Principles of Convolutional Neural Networks

In-depth discussion of the principles and development of convolutional neural networks, including core concepts such as convolutional kernels, convolutional strides, and margin expansion, to help readers comprehensively understand the application of deep learning in the field of computer vision.

0 words Last Updated: 2025-02-18

60. Convolutional Neural Network Construction

This course details how to build convolutional neural network models using TensorFlow and PyTorch, including implementation methods for low-order and high-order APIs, and practices training with the MNIST dataset as an example. Learning this course can master the core principles and implementation techniques of convolutional neural networks.

590 words Last Updated: 2025-02-18

61. Build the LeNet-5 Estimator

This article introduces how to build the classic LeNet-5 convolutional neural network model using the TensorFlow Estimator API, and completes model training and evaluation. Suitable for readers with a basic understanding of deep learning.

16 words Last Updated: 2025-02-18

62. Principles and Practices of Image Classification

This article introduces the application principles and practices of image classification in machine learning and deep learning, including data loading, transfer learning, and cat and dog recognition, and provides methods for visual analysis to help readers deeply understand the core technologies of image classification.

449 words Last Updated: 2025-02-18

63. Transfer Learning for Animal Classification

This course will introduce how to use transfer learning and TensorFlow Keras to complete the animal classification task. By using the MobileNetV2 pre-trained model, you can quickly build an efficient animal recognition model suitable for small dataset scenarios.

48 words Last Updated: 2025-02-18

64. Principle and Construction of Generative Adversarial Networks

This article details the principles and implementation methods of generative adversarial networks, including the adversarial training process of generators and discriminators, and the application of GANs in image generation and other fields. Suitable for readers interested in machine learning and deep learning.

263 words Last Updated: 2025-02-18

65. DCGAN Anime Character Image Generation

This tutorial explains how to generate realistic anime character images using Deep Convolutional Generative Adversarial Networks (DCGAN). Implement the DCGAN network structure using PyTorch and train it on an anime character image dataset to generate realistic anime-style avatars.

228 words Last Updated: 2025-02-18

66. Autoencoder Principle and Construction

This article details the principles and construction methods of autoencoders, including basic autoencoders and denoising autoencoders, and takes the MNIST dataset as an example for practical operations, providing readers with detailed guidance for a comprehensive understanding of autoencoders.

129 words Last Updated: 2025-02-18

67. Convolutional Autoencoder for Image Denoising

Learn to use convolutional autoencoders to complete the image denoising task, and master the application of convolutional neural networks in image processing. This course uses hands-on practice to help you quickly master relevant knowledge points.

32 words Last Updated: 2025-02-18

68. Object Detection Principles and Practices

In-depth discussion of the principles and practices of object detection, including mainstream methods such as the R-CNN family, YOLO and SSD, and Mask R-CNN, providing valuable technical insights for applications such as autonomous driving.

207 words Last Updated: 2025-02-18

69. YOLO Image Object Detection Application

YOLO is a fast and efficient real-time image object detection algorithm. This article details the principles and practical applications of YOLO, helping readers master cutting-edge artificial intelligence technology.

42 words Last Updated: 2025-02-18

Deep Learning Applications (Natural Language Processing)#

70. Principles of Recurrent Neural Networks

Recurrent Neural Networks have achieved great success in natural language processing, speech recognition, and other fields. Understanding their principles and applications will help you master the cutting-edge technologies of deep learning. This course details key knowledge points such as sequence models, simple recurrent neural networks, LSTM, and GRU.

0 words Last Updated: 2025-02-18

71. Construction of Recurrent Neural Networks

This course will teach you how to build and train recurrent neural network models using deep learning frameworks. Content includes the IMDB dataset, word embeddings, simple recurrent neural networks, and LSTM recurrent neural networks. Learn how to use high-order APIs to implement complex recurrent neural network models.

75 words Last Updated: 2025-02-18

72. LSTM Prediction of Stock Prices

Explore the possibility of using machine learning techniques for quantitative trading by building an LSTM network and predicting stock prices. This course provides a detailed hands-on tutorial to help readers master the relevant knowledge and skills.

77 words Last Updated: 2025-02-18

73. Text Classification Principles and Practices

This article details the principles and practices of text classification, including core content such as Chinese text segmentation and feature extraction, and conducts practice on the fake news classification task, providing readers with valuable references for a comprehensive understanding of natural language processing and machine learning applications.

792 words Last Updated: 2025-02-18

74. Fake News Classification with Deep Learning

This article introduces how to use deep learning techniques to complete the practical operation of fake news text classification. Through data preprocessing, feature extraction, and deep neural network modeling steps, the accuracy of fake news classification can be improved, contributing to online information security.

4 words Last Updated: 2025-02-18

75. Natural Language Processing Framework Expansion

This course introduces commonly used natural language processing tools, such as NLTK and PyTorch Flair, helping learners to deeply understand the methods and techniques of natural language processing, and laying a foundation for subsequent deep learning applications.

812 words Last Updated: 2025-02-18

76. Google BERT Pre-training Technology

Google BERT is the most advanced NLP pretraining technique, supporting Chinese and multiple languages, which can be used for tasks such as text classification. This tutorial introduces BERT's knowledge points and provides practical challenges to help readers master the application of BERT.

0 words Last Updated: 2025-02-18

77. Neural Machine Translation and Dialogue Systems

Explore the principles and practices of neural machine translation and conversational systems, and master the application of recurrent neural networks in natural language processing, to contribute to the development of artificial intelligence technology.

1480 words Last Updated: 2025-02-18

Deep Learning Engineering (Model Deployment and Inference)#

78. A Review of Automated Deep Learning

This article details the concepts, goals, and related techniques of automated deep learning, including automated feature engineering, automated model selection, automatic hyperparameter optimization, and neural architecture search, providing a comprehensive automated learning guide for machine learning and artificial intelligence developers.

0 words Last Updated: 2025-02-18

79. Automated Deep Learning Practice

This article introduces the application of the Auto-Keras automated deep learning framework, including tasks such as image classification and text classification, and demonstrates model visualization and AutoML advantages and disadvantages analysis, providing practical guidance for beginners.

274 words Last Updated: 2025-02-18

80. Cactus Aerial Photo Classification and Recognition

Aerial Cactus Identification is a machine learning entry-level competition on Kaggle. This challenge will use the Auto-Keras automated deep learning framework to complete the competition, learning aerial cactus image classification recognition and Auto-Keras practical application.

0 words Last Updated: 2025-02-18

81. Deep Learning Model Inference and Deployment

In-depth discussion of deep learning model inference and deployment, including TensorFlow Serving architecture and ONNX open model format, providing comprehensive guidance for machine learning engineering practice.

293 words Last Updated: 2025-02-18

82. Building an Image Classification Inference Service

Learn how to build a real-time image classification inference API using a pre-trained TensorFlow Keras model, which implements the function of classifying and inferring any image.

0 words Last Updated: 2025-02-18

83. Practice of Deep Learning Cloud Services

Explore deep learning applications in the cloud, covering major cloud providers' ML tools and APIs. Learn to leverage cloud services for efficient ML development.

77 words Last Updated: 2025-02-18

84. Identify VAT Invoices for Cloud Services

This challenge requires calling the value-added tax (VAT) invoice recognition interface provided by Baidu Cloud to complete the task of recognizing VAT invoice images. By independently learning and calling the official API, without using the SDK, complete the recognition of input image links.

9 words Last Updated: 2025-02-18

Reinforcement Learning Fundamentals#

85. Introduction and Examples of Reinforcement Learning

Reinforcement learning is one of the important means of realizing artificial intelligence. This course will comprehensively introduce the basic concepts, algorithm classification, and application cases of reinforcement learning, helping learners deeply understand the cutting-edge technologies of artificial intelligence.

0 words Last Updated: 2025-02-18

86. Implementation of Q-Learning Reinforcement Learning Method

This article details the concept and implementation of the value-based reinforcement learning method, Q-Learning. Through case analysis and algorithm flow explanation, it helps readers deeply understand the application of reinforcement learning in machine learning.

233 words Last Updated: 2025-02-18

87. Implement the Sarsa Learning Algorithm to Exit the Maze

This tutorial details how to use the Sarsa algorithm to implement the process of an agent finding the optimal path in a maze. Through steps such as Q-table initialization, action selection, and behavior feedback, learners can deeply master the core concepts and practical applications of reinforcement learning.

125 words Last Updated: 2025-02-18

Appendix (Machine Learning Mathematical Foundations)#

88. Basic Linear Algebra in Python

This tutorial will comprehensively introduce the basic knowledge of linear algebra in Python, including concepts such as vectors, matrices, determinants, and eigenvalues, and provide practical code examples, helping readers deeply understand the application of linear algebra in machine learning.

473 words Last Updated: 2025-02-18

89. Python Calculus Basics

A Python calculus fundamentals tutorial that explains in detail the required mathematical foundations for machine learning algorithms, such as linear functions, derivatives, partial derivatives, and gradient descent algorithms, along with rich examples, helping readers deeply understand and master the application of calculus in machine learning.

134 words Last Updated: 2025-02-18

90. Basic Probability Theory and Statistics in Python

This course comprehensively introduces the application of probability theory and statistics in machine learning, including core knowledge such as probability formulas, random variables, probability distributions, mathematical expectations, and hypothesis testing, helping you master the theoretical foundations of machine learning.

340 words Last Updated: 2025-02-18

Appendix (Machine Learning Common Tools)#

91. A Concise Guide to Jupyter Notebook

Jupyter Notebook is an essential tool for machine learning and data analysis. This article details its basic functions and usage, helping beginners get started quickly.

66 words Last Updated: 2025-02-18

92. NumPy Numerical Computing Basics

Master NumPy basics for numerical computing in Python. Learn array operations, matrix calculations, and mathematical functions essential for machine learning.

97 words Last Updated: 2025-02-18

93. Basic Pandas Data Processing

Pandas is a very well-known open-source data processing library that can help you quickly read, transform, filter, and analyze datasets. This course will introduce the core functions of Pandas such as basic data types, data reading, selection, deletion, and filling, helping you master the essential skills of data preprocessing.

71 words Last Updated: 2025-02-18

94. Matplotlib 2D Image Plotting Methods

Matplotlib is an open-source plotting library for the Python language, supporting a rich variety of plot types and easy plotting methods. It is a commonly used visualization tool in machine learning and data analysis. This article details the methods and techniques for drawing two-dimensional graphics with Matplotlib.

349 words Last Updated: 2025-02-18

95. Matplotlib 3D Plotting Methods

Matplotlib is an open-source plotting library that supports the Python language, which can help you easily draw various types of 3D graphics, including scatter plots, line graphs, bar graphs, and surface plots. This article details the methods and techniques for drawing 3D graphics with Matplotlib, providing you with professional data visualization solutions.

120 words Last Updated: 2025-02-18

96. Seaborn Data Visualization Basics

Master data visualization with Seaborn, a powerful library built on Matplotlib. Learn to create beautiful statistical graphics for data exploration and analysis.

28 words Last Updated: 2025-02-18

97. SciPy Scientific Computing Basics

SciPy is an important scientific computing library in Python, containing many functional modules such as statistics, optimization, and linear algebra. This course details the basic usage of SciPy, helping you master the core skills of numerical computing and machine learning.

58 words Last Updated: 2025-02-18

Discussion#

If you have any questions about the material, feel free to leave a comment at the bottom of each page. I’ll respond when possible, and you’re welcome to help answer others’ questions too.

No content is perfect - if you spot any errors, typos, or code issues, please point them out in the comments. I’ll verify and correct them as soon as possible.

Acknowledgments#

Listed chronologically:

If you’ve recommended this site, please let me know through comments or email, and I’ll add a Do Follow link to acknowledge your support.

Contributing#

If you’d like to help improve the content or fix errors, please start by creating an issue on GitHub indicating your interest in contributing. Once confirmed, I’ll invite you to join the content collaboration repository.

○ Sharethis article link to your social media, blog, forum, etc. More external links will increase the search engine ranking of this site.

Next

Environmental Overview