“…Understand how various cloud ecosystem components interact (i.e., Amazon S3, EMR, EC2, and so on) Learn how to architect the components of a cloud ecosystem into an end-to-end model pipeline Explore the capabilities and limitations of Spark in building an end-to-end model pipeline Learn to write, publish, deploy, and schedule an ETL process using Spark on AWS using EMR Understand how to create a pipeline that supports model reproducibility and reliability Jason Slepicka is a senior data engineer with Los Angeles based DataScience, where he builds pipelines and data science platform infrastructure. …”

Enlace del recurso

Acceso restringido con credenciales, usuarios UPSA

Tabla de Contenidos: “…Boston Chicago; Dallas; Los Angeles County; New York; Summary; Chapter Four - Strategies for Improving Access and Quality; Strategies to Improve Access; Progress on Access; Strategies to Improve Quality; Tensions and Limitations Inherent in Strategies; Summary; Chapter Five - Sparking and Sustaining Coordination Across Providers and Influencers; Coordination as an Enabler of Improved Access and Quality; Conditions That Foster Coordination; Conditions That Impede Coordination; Advantages of Participating in Coordination Efforts; Drawbacks of Participating in Coordination Efforts; Summary…”

“…Understand how to set-up and manage an enterprise ML model pipeline Learn the common components that make up enterprise ML model pipelines Explore the use and purpose of pipeline tools like Spark, Hive, ModelDB, and Docker Discover the gaps in the Spark ecosystem for maintaining and deploying ML pipelines Learn how to move from creating one-off models to building a reproducible automated pipeline Jason Slepicka is a senior data engineer with Los Angeles based DataScience, where he builds pipelines and data science platform infrastructure. …”