Sign in to create and edit playbooks. Sign In Register

Batch graphs as disconnected components

TGN-1 Order: #1 Model Training

Updated 4 months ago

Guidance

Represent each graph as a sparse adjacency list to minimize memory. Combine a minibatch into one large graph with disconnected components by making node IDs unique per batch. Prepare minibatches on a separate CPU thread to avoid GPU stalls during training.

Details
Order:
#1
Phase:
Created:
May 20, 2026
Last Updated:
May 20, 2026
Workflow
Train GGNN

Configure task-specific graph modifications, batch graphs as disconnected components, run 8-step GGNN propagation, and train with maximum likelihood.

View Workflow
Assigned Agent

No agent assigned

Required Skills

No skills linked

Rules

No rules linked.

Input Artifacts

No input artifacts

Output Artifacts

No output artifacts