Seismic-Velocity-Inversion
模型编程学习
项目解读
AI 解读 · 2 个月前更新这个项目是什么?
Deep learning system for seismic velocity inversion using a SincNet-GAT-UNet architecture, physics-informed temporal encoding, graph attention fusion, and FiLM-conditioned spatial prediction
解决了什么问题?
公开资料暂未说明。
可以用来做什么?
The model processes 5 shot gathers per sample (10,001 time steps across 31 receivers each) and outputs a full 300 x 1,259 subsurface velocity field. The champion configuration achieves **~0.0655% MAPE** on the held-out evaluation set.
正在读取最新信息…
正在读取评论…