Multigraph fusion and classification network using graph neural network
-
Updated
Apr 15, 2022 - Python
Multigraph fusion and classification network using graph neural network
FlyWire FAFB v783 connectome as a ready-to-train PyTorch Geometric dataset: 139,255 neurons, 2.7M edges, 9-class node classification, real CPU baselines (MLP 0.9851 / SAGE 0.9812 / GCN 0.9166)
A few-shot learning approach to forecasting the evolution of the brain connectome.
Local-first context graph and Sync Session platform for human-agent alignment. RESTful API, D3/Three.js visualizations, static JSON contract.
Manage AI memory, rules, and context with a visual dashboard for MCP tools, Claude, ChatGPT, and Cursor
To associate your repository with the brain-map topic, visit your repo's landing page and select "manage topics."