|
所在平台: Coursera |
课程主页: https://www.coursera.org/learn/ibm-rapid-prototyping-watson-studio-autoai
课程评论：没有评论
课程名称：IBM Watson Studio的机器学习快速原型开发 课程概述：当前人工智能领域的一个新趋势是利用自动化技术选择最合适的模型、进行特征工程和通过超参数优化提升模型性能。这种自动化将使模型的快速原型开发成为可能，使数据科学家能够将精力集中在应用领域知识以微调模型上。此课程将带领学习者通过Watson Studio的AutoAI实验工具创建一个端到端的自动化管道，解释IBM Research开发的基础技术。课程重点是使用自动生成的Python笔记本进行操作，学习者将获得两个用例的测试数据集。 该课程面向实践中的数据科学家，虽展示了IBM Watson Studio与AutoAI的自动化AI能力，但不解释机器学习或数据科学的基础概念。 为成功完成课程，学习者需具备以下知识： - 数据科学工作流程 - 数据预处理 - 特征工程 - 机器学习算法 - 超参数优化 - 模型评估指标 - Python及scikit-learn库（包括Pipeline类） 课程大纲： 部分1: 使用Watson Studio AutoAI构建快速原型 描述：本模块中，你将学习AutoAI技术的最新发展，还将熟悉Watson Studio平台，以进行自己的AutoAI实验。在观察AutoAI工具为两个用例构建原型后，你将亲自试用该工具构建额外的原型。 部分2: 自动化数据准备与模型选择 描述：在本模块中，您将了解AutoAI执行的自动化数据准备技术，并有机会在AutoAI生成的Python笔记本中实验不同的数据预处理设置。同时，您将学习自动模型选择的过程，并在数据集上实验不同的模型。 部分3: 自动化特征工程与超参数优化 描述：在本模块中，您将学习自动化特征工程的算法，并进行一些探索性数据分析，以了解算法为何执行特定的特征转换。您还将学习优化超参数的复杂方法，并在数据集上探索使用AutoAI生成的Python笔记本进行超参数调优。 部分4: AutoAI生成解决方案的评估与部署 描述：在本模块中，您将使用AutoAI工具计算的不同评估指标来评估原型。您还将利用Watson机器学习API部署原型以进行测试。
Part: 1
Title:Building a Rapid Prototype with Watson Studio AutoAI
Description:In this module, you'll learn about the developing landscape of AutoAI technologies. You'll also become familiar with the Watson Studio platform in order to be able to perform your own AutoAI Experiments. After observing the AutoAI tool build prototypes for two use cases, you will try out the tool for yourself to build additional prototypes.
Part: 2
Title:Automated Data Preparation and Model Selection
Description:In this module, you will learn about the automated data preparation techniques performed by AutoAI and get a chance to experiment with different settings for data preprocessing in the AutoAI-generated Python notebook. You'll also learn about the procedure for automated model selection and experiment using different models on the datasets.
Part: 3
Title:Automated Feature Engineering and Hyperparameter Optimization
Description:In this module, you will learn about the algorithm for automated feature engineering and perform some exploratory data analysis to try to understand why the algorithm performed particular feature transformations. You'll also learn about sophisticated methods for optimizing hyperparameters and explore hyperparameter tuning on the datasets using the AutoAI-generated Python notebook.
Part: 4
Title:Evaluation and Deployment of AutoAI-generated Solutions
Description:In this module, you will evaluate prototypes using the different evaluation metrics calculated by the AutoAI tool. You will also deploy the prototype for testing using the Watson Machine Learning API.
An emerging trend in AI is the availability of technologies in which automation is used to select a best-fit model, perform feature engineering and improve model performance via hyperparameter optimization. This automation will provide rapid-prototyping of models and allow the Data Scientist to focus their efforts on applying domain knowledge to fine-tune models. This course will take the learner through the creation of an end-to-end automated pipeline built by Watson Studio’s AutoAI experiment tool, explaining the underlying technology at work as developed by IBM Research. The focus will be on working with an auto-generated Python notebook. Learners will be provided with test data sets for two use cases. This course is intended for practicing Data Scientists. While it showcases the automated AI capabilies of IBM Watson Studio with AutoAI, the course does not explain Machine Learning or Data Science concepts. In order to be successful, you should have knowledge of: Data Science workflow Data Preprocessing Feature Engineering Machine Learning Algorithms Hyperparameter Optimization Evaluation measures for models Python and scikit-learn library (including Pipeline class)
