Skip to content

Conversation

@Le-soleile
Copy link
Contributor

@Le-soleile Le-soleile commented Nov 7, 2025

PR Category
Dataset adaptation

PR Types
Add adapters

Description

在CHGNet模型上测试
示例 bash: python interatomic_potentials/train.py -c interatomic_potentials/configs/chgnet/chgnet_oc20_s2ef_energy.yaml
image
image
image

@paddle-bot
Copy link

paddle-bot bot commented Nov 7, 2025

Thanks for your contribution!

"排查建议:\n"
" 1) 检查配置的 label_names 与分片列是否一致\n"
" 2) 暂时将 filter_unvalid 设为 False 以查看原始样本计数\n"
" 3) 仅保留 1 个分片进行快速验证\n"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

请使用英文

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

请参考已有的数据集dataset的文件格式,url写在dataset里面

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

参考mp的格式

@leeleolay
Copy link
Collaborator

最终的数据集加载文件中用到pyarror的话,辛苦在requirement中辛苦添加相关命令

@leeleolay
Copy link
Collaborator

这个数据建议使用chrgnet

@Le-soleile
Copy link
Contributor Author

建议已经全部修改完毕💕

@Le-soleile
Copy link
Contributor Author

导师您看这个还有问题嘛

@Le-soleile Le-soleile changed the title 【PPMat Dataset No.6】OC20_MEGNet 【PPMat Dataset No.6】OC20_CHGNet Nov 29, 2025
@Le-soleile
Copy link
Contributor Author

导师您看这个还有问题嘛

@leeleolay

@leeleolay
Copy link
Collaborator

每次运行都要重新构建graph,这里的逻辑有点问题

@leeleolay
Copy link
Collaborator

建议不要分chunk构建graph,一个进度条的方式来显示完整的graph构建过程

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

contributor External developers

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants