|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/automated-machine-learning-auto-ml-tpot-h2o-auto-keras/
课程评论：没有评论
课程名称：自动化机器学习 - AutoML，TPOT，H2O，AutoKeras 课程概述：加入这门全面的课程，深入探讨自动化机器学习（AutoML）技术。课程将介绍多种强大的工具，包括TPOT、AutoML、AutoKeras和H2O。您将学习如何比较和对比堆叠的机器学习模型与自动化模型，获得在解决优化问题时的有效性洞察。此外，我们还将进行五个实践练习，包括： 1. 使用信用卡欺诈数据集进行AutoML：在此练习中，您将利用AutoML技术自动构建和优化机器学习模型，以检测信用卡欺诈。AutoML算法将自动探索多种模型、特征工程技术和超参数配置，以识别最有效的解决方案。 2. 在MNIST数据集上使用AutoKeras：MNIST是一个经典的数据集，常用于手写数字识别。通过AutoKeras，您将自动构建和调整深度神经网络，以准确分类MNIST中的手写数字。 3. 使用TPOT进行保险预测：TPOT（基于树的管道优化工具）是一种自动发现和优化机器学习管道的AutoML工具。在此练习中，您将应用TPOT来预测与保险相关的结果，例如保险索赔或客户行为。 4. 使用H2O进行客户流失预测：客户流失预测涉及预测客户是否会停止使用某项服务或产品。通过H2O，您将构建预测模型，以识别潜在的流失客户。 5. 使用H2O进行销售预测：销售预测涉及根据历史数据和其他相关因素预测未来销售。在此练习中，您将利用H2O开发销售预测模型。 无论您是希望优化工作流程的资深数据科学家，还是渴望掌握最新机器学习进展的新手，这门课程都为您提供了一个实用且富有洞察力的自动化机器学习之旅。
Join this comprehensive course as we delve into the Automated Machine Learning (AutoML) Techniques. Throughout the program, we'll explore a variety of powerful tools including TPOTs, AutoML, AutoKeras, and H2O.You'll learn to compare and contrast Stacked Machine Learning Models with Automated counterparts, gaining valuable insights into their efficacy for solving optimization problems. Additionally, we will work on 5 excercises which includes:AutoML using Credit Card Fraud dataset: In this exercise, you'll leverage AutoML techniques to automate the process of building and optimizing machine learning models to detect credit card fraud. AutoML algorithms will automatically explore various models, feature engineering techniques, and hyperparameter configurations to identify the most effective solution for detecting fraudulent transactions within credit card dataAutoKeras on MNIST data: MNIST is a classic dataset commonly used for handwritten digit recognition. With AutoKeras, a powerful AutoML library specifically designed for deep learning tasks, you'll automate the process of building and tuning deep neural networks for accurately classifying handwritten digits in the MNIST dataset.TPOT for Insurance Predictions: TPOT (Tree-based Pipeline Optimization Tool) is an AutoML tool that automatically discovers and optimizes machine learning pipelines. In this exercise, you'll apply TPOT to the task of predicting insurance-related outcomes, such as insurance claims or customer behavior.Churn Prediction using H2O: Churn prediction involves forecasting whether customers are likely to stop using a service or product. With H2O, an open-source machine learning platform, you'll build predictive models to identify potential churners within a customer base.Sales Prediction using H2O: Sales prediction involves forecasting future sales based on historical data and other relevant factors. In this exercise, you'll utilize H2O to develop predictive models for sales forecasting.Whether you're a seasoned data scientist looking to streamline your workflow or a newcomer eager to grasp the latest advancements in machine learning, this course offers a practical and insightful journey into the world of Automated Machine Learning.
