Skip to content

Conversation

@GRCaaaa
Copy link
Contributor

@GRCaaaa GRCaaaa commented Oct 23, 2025

PR types

PR changes

Describe

@paddle-bot
Copy link

paddle-bot bot commented Oct 23, 2025

Thanks for your contribution!

@GRCaaaa
Copy link
Contributor Author

GRCaaaa commented Oct 23, 2025

数据和模型链接为通过网盘分享的文件:saved_models_ppsci等2个文件
链接: https://pan.baidu.com/s/1ot-7XQo304GZevBiXKT4hg?pwd=yvbk 提取码: yvbk

mkdocs.yml Outdated
Comment on lines 225 to 230
# setup_commands:
# - import sys
# - sys.path.append("../")
paths: [../ppsci]
selection:
new_path_syntax: true
# selection:
# new_path_syntax: true
Copy link
Collaborator

Choose a reason for hiding this comment

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

请不要修改无关代码

x = self.features(x)
x = paddle.flatten(x, 1)
if self.classifier is None:
# 动态推断flatten后的特征维度
Copy link
Collaborator

Choose a reason for hiding this comment

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

代码中的注释请统一使用英文

Comment on lines 127 to 129
def vgg(config="vgg16", pretrained=False, pretrained_path=None, **kwargs):
model = VGG(config=config, **kwargs)
return model
Copy link
Collaborator

Choose a reason for hiding this comment

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

这个函数无用,删除

Copy link
Collaborator

Choose a reason for hiding this comment

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

注释统一使用英文

### 环境要求

```bash
pip install paddlepaddle ppsci numpy pandas scikit-learn tqdm
Copy link
Collaborator

Choose a reason for hiding this comment

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

这个案例只需要使用cpu的paddle运行吗?不需要装-gpu版的?

@leeleolay
Copy link
Contributor

@GRCaaaa
Copy link
Contributor Author

GRCaaaa commented Oct 23, 2025

由于这个后续涉及项目后面进行了一步模型集成学习,从而保留了较多的模型,但是这一步不是文章中的关键,从而舍弃了。由于训练数据较小,采取cpu即可,从而我直接设置的cpu,不过改成cuda是可以运行的。

@leeleolay
Copy link
Contributor

参考其他案例 整理一下文档,提供运行的命令

@HydrogenSulfate HydrogenSulfate merged commit 5350fc9 into PaddlePaddle:develop Oct 31, 2025
2 of 3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants