한국어
로그인
Tensorflow를 사용한 기계 학습 알고리즘(Python 레벨 5)
보고계신 지문은 자동 번역 되었습니다
수업 소개
This comprehensive Machine Learning Fundamentals course covers essential concepts and practical applications across ten structured lessons. Beginning with an introduction to TensorFlow and basic machine learning principles, students progress through topics such as linear regression, model tuning strategies, K-NN clustering, decision forests, and neural networks, including feed-forward networks and convolutional neural networks (CNNs). Each lesson incorporates hands-on projects using datasets...
10 lessons//10 Weeks
Week 1Lesson 1Intro To TensorFlowObjectives: Learn the basics of machine learning What is learning? History of machine learning Regression vs classification Introduce TensorFlow Installing TensorFlow What is a Tensor? Low-level API and high-level API How do Tensors perform computations? Tensorflow variables Project 1: Perform basic computations with TensorFlow Matrix addition, matrix multiplication using tensors in TensorFlowWeek 2Lesson 2Linear RegressionThis project aims to teach linear regression fundamentals, including hypothesis setup with the equation 𝑦 = 𝑚 𝑥 + 𝑏 y=mx+b, where 𝑚 m represents weights and 𝑏 b denotes biases. Participants will learn to compute these parameters using gradient descent and understand the mean squared error cost function. Using TensorFlow and NumPy, they will implement a linear regression model, iteratively refining weights and biases over multiple epochs and visualizing model performance.Week 3Lesson 3Model TuningObjectives Why do we need to tune models? Convergence problems Preparation Strategies for tuning models Grid Search Random Search Data transformation Project 3: Tune the linear regression model to produce better results Implement Grid Search Implement Random Search Implement a data transformationWeek 4Lesson 4K-NN ClusteringObjectives: How do we solve classification problems? Problems with classification Strategies and models to use K-NN Algorithms How do clustering algorithms work? How does the K-NN algorithm perform classification How do we measure distance? What is the effect of the K hyperparameter? Project 4: MNIST Classification with K-NN Import the dataset and process it Implement the K-NN algorithm Visualize the results Tune the model for better resultsWeek 5Lesson 5Decision ForestsObjectives: What are Ensembles and Decision Forests? How do decision trees work and their shortcomings? What is an ensemble and how do they make predictions? What is a decision forest? Tuning the forest What is Gradient boost? What hyperparameters does a Decision Forest have? Project 5: Fraud Detection using Decision Forests Preprocess data Create and train model Visualize results Test other optimization methods to produce the best modelWeek 6Lesson 6Intro To Neural NetworksObjectives Cover the basics of Neural Networks Neural Network playground Nodes, weights, bias, Linear combination Input Layer Hidden Layer Output Layer Explain data manipulation and ways to view data Length of dataset Shape of dataset Preprocess images Verify the images are in the correct format Project 6: Fashion MNIST dataset preparation Import the dataset Prepare dataset for machine learningWeek 7Lesson 7Neural Networks ContinuedObjectives What are Feed-forward neural networks? Activation functions (Relu) Loss functions (Cross-entropy) Optimizer functions (ADAM) Feed forward Backpropagation Project 7: Fashion MNIST Flatten dataset for model Create and train the neural network Visualize the resultsWeek 8Lesson 8Convolutional Neural NetworkObjectives: Explain Convolutional Neural networks The differences with feed-forward neural networks What is a convolutional layer? When to use CNN? Project 8: CIFAR10 database classification Prepare the dataset Create the CNN and train it Visualize the results - Training CurvesWeek 9Lesson 9CNN Tuning & Final Project WorkshopRegularization Why do we need regularization? What is L2 regularization? What is dropout regularization? Mini Project 9: Tune CIFAR10 model Implement dropout and L2 regularization View results Final Project Workshop Discuss ideas for a project Help students find a dataset Begin work on projectsWeek 10Lesson 10Final ProjectFinal Projects Students will add finishing touches to their projects Students will take turns presenting their work Final Project Ideas Face Identification Character recognition Create ensembles of models to tackle a difficult challenge Analyze the effects of different hyperparameters for models
- This Machine Learning Fundamentals course covers TensorFlow introduction, basic principles, linear regression, model tuning, K-NN clustering, decision forests, and neural networks like CNNs.
- Through hands-on projects with MNIST and CIFAR10 datasets, students implement algorithms, optimize models, and present final projects on topics including image classification and hyperparameter optimization.
수업 외 주당 2 - 4 시간
Projects
빈도: 1-2 throughout the class피드백: 포함됨세부 내용: Projects are not mandatory but we strongly encourage students to complete them.Assessment
빈도: 포함됨세부 내용:
리뷰
그룹 수업
10 회 수업에
₩259
10주 동안 주당 1회
60분
8 명의 학생이 수업을 완료함
실시간 화상 수업
연령: 13-18
수업당 학습자 4-8 명