arXiv cs.AI· Eloi Campagne (CB), Yvenn Amara-Ouali (LMO, CELESTE), Yannig Goude (EDF R\&D), Argyris Kalogeratos (CB, ENS Paris Saclay)·· 4 小时前AI 评分36
GraphToolbox：面向图神经网络电力预测可控基准测试的开源可配置框架
GraphToolbox: A Configurable Framework for Controlled Benchmarking of Graph Neural Network Forecasting
AI 导读
GraphToolbox 是一个开源 Python 框架，基于 PyTorch Geometric 将图构建、模型选择、训练、在线专家聚合与可解释性等 GNN 电力预测实验环节统一到配置驱动流水线中。
正文 · 原文
Abstract:Electricity forecasting often involves spatially related signals observed over regions, substations, and feeders, and Graph Neural Networks (GNNs) represent these relations through a graph. Building a complete GNN forecasting experiment is nonetheless laborious, because graph construction, model selection, training, aggregation, and interpretation often require separate tools. We present GraphToolbox, an open-source Python framework that unifies these stages in one configuration-driven pipeline built on PyTorch Geometric. It offers data-driven graph construction, a single interface for most graph convolution operators of PyTorch Geometric and for recurrent graph cells, online expert aggregation, forecasting interpretability, and significance testing on forecasts. We use GraphToolbox to run a controlled benchmark: we compare architectural and modeling choices for French electricity load and net-load forecasting at regional and national levels under a common experimental protocol. On load, most evaluated graph convolutions outperform additive, recurrent, and foundation-model baselines. Pooled XGBoost, trained on the same inputs, has a lower RMSE than the best graph models, although the difference is not significant. Online expert aggregation and daily model updates improve accuracy further. On net-load, graph models outperform additive, recurrent, and XGBoost baselines, with further gains from forecasting the physical components separately. Statistical tests do not identify a single best convolution for load forecasting.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.24609 [cs.LG] |
| (or arXiv:2609.24609v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2609.24609 arXiv-issued DOI via DataCite |
Submission history
From: Eloi Campagne [view email] [via CCSD proxy]
[v1]
Mon, 21 Sep 2026 13:58:35 UTC (1,066 KB)
[v2]
Tue, 10 Nov 2026 14:55:49 UTC (1,354 KB)
来源：arXiv cs.AI · arxiv.org