Caesar AI Atlas

Pipeline

Caesar AI Atlas Definition

Un pipeline est une séquence ordonnée d’étapes de traitement qui fait circuler des données ou des artefacts dans un workflow d’IA ou d’apprentissage automatique. Il peut inclure des étapes d’ingestion, de transformation, d’entraînement, d’évaluation, de déploiement et de surveillance.

Other Definitions

Pipeline Source

All of the operations needed to fit a model to a data set. A pipeline consists of data import, transformation, featurization, and learning steps. Once a pipeline is trained, it turns into a model.

Pipeline Source

The infrastructure surrounding a machine learning algorithm. A pipeline includes gathering the data, putting the data into training data files, training one or more models, and exporting the models to production. See ML pipelines in the Managing ML Projects course for more information.

Pipeline Source

ML pipelines are portable and scalable ML workflows that are based on containers. For more information, see Introduction to Vertex AI Pipelines.

Also Referenced In

Related Terms