|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/amazon-machine-learning-engineer-certificate-practice-exam/
课程评论：没有评论
课程名称：亚马逊 - 机器学习工程师证书实践考试 课程概述：完成AWS认证机器学习工程师-助理课程的专业人士将具备创建、训练、实施和优化机器学习模型所需的知识和能力，并能够利用AWS服务。该课程专为未来的机器学习工程师设计，填补了云计算机器学习理论与实践之间的知识空白。 课程内容从数据工程入手，涵盖了实时数据处理工具，如Amazon Kinesis，以及存储选项，如Amazon S3。参与者将学习特征工程、数据清洗及转换活动，利用AWS Glue和SageMaker Data Wrangler等工具进行操作。接着，课程深入模型构建，指导学习者选择最佳算法、调整超参数，并通过ROC-AUC和F1评分等指标评估模型。学生将通过实际实验，获取使用SageMaker内置算法和框架（如TensorFlow和PyTorch）训练和优化模型的经验。 部署阶段的重点是MLOps，介绍了工作流自动化工具、使用SageMaker终端进行模型部署，以及CI/CD管道集成以实现可扩展性。此外，参与者还将学习版本控制、成本优化技巧以及安全实施机器学习系统的最佳实践。 最后，课程涵盖监控和维护，强调使用AWS产品（如SageMaker Model Monitor）进行持续改进，管理模型性能以及识别数据漂移。
Professionals who complete the AWS Certified Machine Learning Engineer-Associate course will have the knowledge and abilities necessary to create, train, implement, and enhance machine learning models with AWS services. Designed for future machine learning engineers, this program fills the knowledge gap between theory and practice in cloud-based machine learning.Starting with data engineering, the course covers real-time data processing tools like Amazon Kinesis and storage options like Amazon S3. Participants will gain knowledge of feature engineering, data cleansing, and transformation activities utilizing tools like AWS Glue and SageMaker Data Wrangler.The course then delves into model construction, instructing learners on how to choose the best algorithms, adjust hyperparameters, and assess models using metrics such as ROC-AUC and F1 score. Students will get experience training and refining models using SageMaker's built-in algorithms and frameworks, such as TensorFlow and PyTorch, through practical experiments.MLOps are the focus of the deployment phase, which also introduces tools for workflow automation, model deployment with SageMaker endpoints, and CI/CD pipeline integration for scalability. Additionally, version control, cost optimization techniques, and best practices for implementing safe machine learning systems will be taught to participants.Last but not least, the course covers monitoring and maintenance, emphasizing the use of AWS products such as SageMaker Model Monitor for continuous improvement, managing model performance, and identifying data drift.
