GraphForge

Relation/topology learning core (GNN/GAT/GraphTransformer)

Product category: Graph learning

Applicable platforms: PyG/DGL, PyTorch

Technical affiliation: Graph ML, Message Passing

Programming language affiliation: Python

Tags: #GraphML #GNN Product

Type: Graph learning

GraphForge provides a graph modeling core for "who is related to whom" and "how structure affects prediction/policy", supporting static/dynamic graphs, sampling/subgraph training, message passing, attention, and graph readout. Fused with time series/multimodal, it can build spatio-temporal graph models to capture diffusion/adjacent influences; combined with decision, it can be used as policy structural context or feasible domain constraint prior. It also provides graph explanation (node/edge contribution) and visualization tools, helping R&D and business understand model behavior from a structural perspective, enhancing interpretability and controllability.

Delivery method: Instant digital download after purchase
License: Single-user commercial license
Usage limit: One-time use
Support: Technical documentation provided in the delivery file; no human technical support included