Directed acyclic graph

As explained in Machine Learning System Design

A directed acyclic graph is a directed graph whose edges have no cycle, so its nodes can be arranged in an evaluation order.

A directed acyclic graph (DAG) models an ordered pipeline: each step receives input, performs an action, and passes its output to a later step without forming a cycle. section 10.1.1

In machine learning, a training pipeline is a higher-level DAG that turns a full dataset and optional metadata into a trained model. An inference pipeline is a lower-level DAG that turns raw data into predictions. section 10.1.1

A machine-learning pipeline

A typical image-classification DAG can connect image reading, normalization, resizing, and model execution. A training DAG commonly connects data fetching, preprocessing, and model training. section 10.1.1

Figure 10.1 A DAG scheme representing a training pipeline

Use the DAG figure to visualize how training steps connect from data inputs to a trained model. section 10.1.1

How Machine Learning System Design covers it

How deep it goes

Defines DAGs through pipeline inputs, actions, and outputs, then distinguishes training from inference without detailing graph algorithms.

Examples

Uses image reading, normalization, resizing, model execution, data fetching, preprocessing, and model training as pipeline steps.

What it assumes

Assumes ordered steps with inputs and outputs, then clarifies how those steps form machine-learning pipeline abstractions.

What you can do after

Evaluate machine-learning workflows by distinguishing higher-level training DAGs from lower-level inference DAGs.

How it connects

Connects DAGs with machine-learning pipelines, preprocessing, feature engineering, model training, and predictions.

liveBookmanning.com

Read Machine Learning System Design on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Causal Inference for Data ScienceA directed acyclic graph is a graph with directed edges and no directed cycles, used in causal inference to represent variables and their causal relationships.
• 

  liveBookmanning.com
  Designing Deep Learning Systems: A guide for software engineersA directed acyclic graph is a directed graph whose edges encode dependencies and whose directed paths contain no cycles.
• 

  liveBookmanning.com
  Data Pipelines with Apache Airflow, Second EditionA directed acyclic graph is a directed graph with no cycles, often used to represent workflow tasks and their dependencies.
• 

  liveBookmanning.com
  Causal AIDAG is a graph class representing a directed acyclic graph and operations on its nodes and edges.
• 

  liveBookmanning.com
  Terraform in Depth: Infrastructure as Code with Terraform and OpenTofuA directed acyclic graph is a directed graph whose edges encode dependencies and whose vertices cannot be reached again by following directed paths.
• 

  liveBookmanning.com
  Snowflake Data EngineeringA directed acyclic graph is a directed graph whose paths contain no circular dependencies.
• 

  liveBookmanning.com
  Architecting an Apache Iceberg LakehouseA directed acyclic graph is a graph of nodes and directed dependencies with no cycles, commonly representing workflow tasks and their execution order.
• 

  liveBookmanning.com
  Delivering Machine Learning Projects: From design to deploymentA directed acyclic graph is a directed network of nodes and edges with no directed cycle.
• 

  liveBookmanning.com
  Distributed Machine Learning PatternsA directed acyclic graph is a directed graph with no path that returns to an earlier vertex, so its dependencies have no cycles.
• 

  liveBookmanning.com
  Natural Language Processing in Action, Second EditionA directed acyclic graph is a graph whose edges have directions and that contains no directed cycles.
• 

  liveBookmanning.com
  Graph Neural Networks in ActionA directed acyclic graph is a directed graph containing no cycles.
• 

  liveBookmanning.com
  Machine Learning Platform EngineeringA directed acyclic graph is a graph whose directed edges connect nodes without forming any cycle.
• 

  liveBookmanning.com
  Inside AIA directed acyclic graph is a directed graph containing no path that returns to its starting node.
• 

  liveBookmanning.com
  Financial AI in PracticeA directed acyclic graph is a directed graph with no directed cycles, often used to represent task dependencies.

Compare how 15 books explain directed acyclic graph →

Related concepts

• acyclic graphkind of · An acyclic graph is a graph that contains no cycles.
• network topologyincludes · Network topology is the structural arrangement and connectivity relationships among the components of a network or system.
• potential outcomes frameworkequivalent to · The potential outcomes framework defines causal effects by comparing outcomes that would result for a unit under alternative treatments or interventions.
• Apache Airflowused by · Apache Airflow is an open-source platform for authoring, scheduling, and monitoring programmatically defined workflows, commonly represented as directed acyclic graphs.
• graphkind of · A graph is a mathematical structure consisting of vertices connected by edges that represent relationships or connections.
• d-separationused by · D-separation is a graphical criterion for determining whether a conditioning set blocks all relevant paths between variables in a graph, implying conditional independence under the graph’s assumptions.

This explainer was generated by Manning's knowledge graph from the text of Machine Learning System Design. The section links point to the parts of the book it draws on.