search

echowve / meshGraphNets_pytorch

PyTorch implementations of Learning Mesh-based Simulation With Graph Networks
Apache License 2.0
152 stars 33 forks source link

Rollout problem #5

Closed fycAIwj closed 1 year ago

fycAIwj commented 1 year ago

Hello dear author, I would like to ask why in the code for prediction, the true value of the next moment is used as the input for rolling prediction, instead of using the predicted value as the input for rolling prediction.

echowve commented 1 year ago

The replaced value represents the boundary conditions of the flow field and is known beforehand.

fycAIwj commented 1 year ago

Dear Author Thanks for your reply, your code is very useful.

发件人：Longxiang Jiang @.> 发送日期：2023-07-07 17:56:51 收件人：echowve/meshGraphNets_pytorch @.> 抄送人：fycAIwj @.>,Author @.> 主题：Re: [echowve/meshGraphNets_pytorch] Rollout problem (Issue #5)

The replaced value represents the boundary conditions of the flow field and is known beforehand. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>