|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/dp-100-designing-and-implementing-data-science-solution-2023/
课程评论：没有评论
**DP-100: 设计与实施数据科学解决方案 2023 - 课程总结** 本课程旨在帮助您为“Microsoft 认证：Azure 数据科学家助理”考试做好充分准备。通过模拟真实考试场景的练习题，您将深入理解 Azure 机器学习的核心概念和实际应用。 **课程重点涵盖：** * **设计与准备机器学习解决方案 (20-25%)** * 确定工作负载的计算要求 * 描述模型部署需求 * 选择合适的开发方法 * 管理 Azure 机器学习工作区（创建、配置 Git 集成） * 管理 Azure 机器学习中的数据（存储资源、数据存储、数据资产） * 管理实验计算（计算目标、环境配置、Spark 池） * **探索数据与训练模型 (35-40%)** * 使用数据资产和数据存储探索和整理数据（包括 Apache Spark） * 使用 Azure 机器学习设计器（创建训练管道、使用自定义组件） * 进行模型评估（包括负责任 AI 指南） * 使用自动化机器学习（用于表格数据、计算机视觉、NLP） * 优化训练选项（预处理、算法、超参数调整） * 使用 Notebooks 进行自定义模型训练（Python SDK v2，MLflow 跟踪） * 使用计算实例进行模型训练和调试 * **为模型部署做准备 (20-25%)** * 运行模型训练脚本（作业配置、计算、数据消费） * 使用 MLflow 记录指标和日志进行故障排除 * 配置作业环境和参数 * 创建和管理训练管道（组件、数据传递、计划、监控） * 包裝模型（MLflow 模型输出、框架选择） * 评估模型（负责任 AI 指南） * **部署与再训练模型 (10-15%)** * 部署模型（在线和批量端点） * 测试部署的模型 * 应用 MLOps 实践（自动化触发器、事件驱动再训练） **目标受众：** 具备集成、转换和整合来自不同结构化和非结构化数据系统数据的专业知识，并将数据构建成适合机器学习解决方案的学员。熟悉 SQL、Python 或 Scala 等数据处理语言，并理解并行处理和数据架构模式。 **祝您考试顺利！**
Are you aiming to achieve the prestigious "Microsoft Certified: Azure Data Scientist Associate" certification? Look no further! Our comprehensive practice test is specially curated to test your knowledge and prepare you with 100% confidence to ace the DP-100 examination.Designed with utmost care, the questions in this practice test are either directly sourced from Azure Documentation or represent real-world data engineering scenarios. This ensures that you are well-prepared for the challenges you may encounter during the exam.Each question is accompanied by detailed explanations and links to the corresponding Microsoft documentation, where the concept or scenario was framed. By taking this practice test, you will not only gain a deep understanding of the subject matter but also get accustomed to the format of the actual DP-100 exam.Our practice test is regularly updated to reflect any changes in the testing areas by Microsoft. You can rest assured that you are practicing with the most relevant and up-to-date content, giving you a competitive edge in the certification process.The objectives covered in this course are:Design and prepare a machine learning solution (20-25%)Design a machine learning solutionDetermine the appropriate compute specifications for a training workloadDescribe model deployment requirementsSelect which development approach to use to build or train a modelManage an Azure Machine Learning workspaceCreate an Azure Machine Learning workspaceManage a workspace by using developer tools for workspace interactionSet up Git integration for source controlManage data in an Azure Machine Learning workspaceSelect Azure Storage resourcesRegister and maintain datastoresCreate and manage data assetsManage compute for experiments in Azure Machine LearningCreate compute targets for experiments and trainingSelect an environment for a machine learning use caseConfigure attached compute resources, including Apache Spark poolsMonitor compute utilizationExplore data and train models (35-40%)Explore data by using data assets and data storesAccess and wrangle data during interactive developmentWrangle interactive data with Apache SparkCreate models by using the Azure Machine Learning designerCreate a training pipelineConsume data assets from the designerUse custom code components in designerEvaluate the model, including responsible AI guidelinesUse automated machine learning to explore optimal modelsUse automated machine learning for tabular dataUse automated machine learning for computer visionUse automated machine learning for natural language processing (NLP)Select and understand training options, including preprocessing and algorithmsEvaluate an automated machine learning run, including responsible AI guidelinesUse notebooks for custom model trainingDevelop code by using a compute instanceTrack model training by using MLflowEvaluate a modelTrain a model by using Python SDKv2Use the terminal to configure a compute instanceTune hyperparameters with Azure Machine LearningSelect a sampling methodDefine the search spaceDefine the primary metricDefine early termination optionsPrepare a model for deployment (20-25%)Run model training scriptsConfigure job run settings for a scriptConfigure compute for a job runConsume data from a data asset in a jobRun a script as a job by using Azure Machine LearningUse MLflow to log metrics from a job runUse logs to troubleshoot job run errorsConfigure an environment for a job runDefine parameters for a jobImplement training pipelinesCreate a pipelinePass data between steps in a pipelineRun and schedule a pipelineMonitor pipeline runsCreate custom componentsUse component-based pipelinesManage models in Azure Machine LearningDescribe MLflow model outputIdentify an appropriate framework to package a modelAssess a model by using responsible AI guidelinesDeploy and retrain a model (10-15%)Deploy a modelConfigure settings for online deploymentConfigure compute for a batch deploymentDeploy a model to an online endpointDeploy a model to a batch endpointTest an online deployed serviceInvoke the batch endpoint to start a batch scoring jobApply machine learning operations (MLOps) practicesTrigger an Azure Machine Learning job, including from Azure DevOps or GitHubAutomate model retraining based on new data additions or data changesDefine event-based retraining triggersCandidates for this exam should have subject matter expertise integrating, transforming, and consolidating data from various structured and unstructured data systems into a structure that is suitable for building Machine Learning solutions, alongside with the knowledge of data processing languages such as SQL, Python, or Scala, and they need to understand parallel processing and data architecture patterns."All the best for your exam"
