Skip to content

Conversation

Xuxuanang
Copy link
Contributor

缺失映射文档补充
os.environ.get
setuptools.setup
torchvision.transforms.InterpolationMode.BICUBIC
torchvision.transforms.InterpolationMode.BILINEAR
torchvision.transforms.InterpolationMode.BOX
torchvision.transforms.InterpolationMode.HAMMING
torchvision.transforms.InterpolationMode.LANCZOS
torchvision.transforms.InterpolationMode.NEAREST
torchvision.transforms.InterpolationMode.NEAREST_EXACT

Copy link

paddle-bot bot commented Nov 13, 2024

感谢你贡献飞桨文档,文档预览构建中,Docs-New 跑完后即可预览,预览链接:http://preview-pr-6962.paddle-docs-preview.paddlepaddle.org.cn/documentation/docs/zh/api/index_cn.html
预览工具的更多说明,请参考:飞桨文档预览工具

os.environ.get(key, value)
```

Paddle 无此 API,需要组合实现。
Copy link
Collaborator

Choose a reason for hiding this comment

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

需要说明一下,该API一般情况下与Paddle无关,仅在torch分布式相关的深度学习用法里才需要转写。并且介绍下相应的分布式用法功能。

setuptools.setup(*args, **attrs)
```

Paddle 无此 API,需要组合实现。
Copy link
Collaborator

Choose a reason for hiding this comment

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

需要说明一下,该API一般情况下与Paddle无关,仅在torch分布式相关的深度学习用法里才需要转写。并且介绍下相应的分布式用法功能。


| PyTorch | PaddlePaddle | 备注 |
| ------------- | ------------ | ------------------------------------------------------ |
| function | function | 模型前向传播的部分连续的层函数组成的序列。 |
Copy link
Collaborator

Choose a reason for hiding this comment

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

这个看起来和上面的函数签名不太符合

'bicubic'
```

两者功能一致,无参数。
Copy link
Collaborator

Choose a reason for hiding this comment

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

简述里写一下 枚举对应到字符串吧

'bilinear'
```

两者功能一致,无参数。
Copy link
Collaborator

Choose a reason for hiding this comment

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

简述里写一下 枚举对应到字符串吧

'hamming'
```

两者功能一致,无参数。
Copy link
Collaborator

Choose a reason for hiding this comment

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

简述里写一下 枚举对应到字符串吧

'lanczos'
```

两者功能一致,无参数。
Copy link
Collaborator

Choose a reason for hiding this comment

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

简述里写一下 枚举对应到字符串吧

'nearest'
```

两者功能一致,无参数。
Copy link
Collaborator

Choose a reason for hiding this comment

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

简述里写一下 枚举对应到字符串吧

'nearest_exact'
```

两者功能一致,无参数。
Copy link
Collaborator

Choose a reason for hiding this comment

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

简述里写一下 枚举对应到字符串吧

@@ -300,7 +300,8 @@ def get_meta_from_diff_file(
if title_match:
mapping_type = title_match["type"].strip()
src_api = unescape_api(title_match["src_api"].strip())

if src_api == "torch.pi":
Copy link
Collaborator

Choose a reason for hiding this comment

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

可以加个白名单列表,后续好维护

setuptools.setup(*args, **attrs)
```

Paddle 无此 API,需要组合实现。该 API 一般情况下与 Paddle 无关,仅在 torch 分布式相关的深度学习用法里才需转写,用来构建一个包含自定义扩展(如 C++ )的 PyTorch 包。
Copy link
Collaborator

Choose a reason for hiding this comment

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

这个不是torch分布式

| function | function | 模型前向传播的部分连续的层函数组成的序列。 |
| preserve_rng_state | preserve_rng_state | 是否保存前向的 rng。 |
| use_reentrant | use_reentrant | recompute 的实现方式。 |
| determinism_check | - | 控制是否在反向传播时检查操作的确定性, Paddle 无此参数,暂无转写方式。 |
Copy link
Collaborator

Choose a reason for hiding this comment

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

这个只是做一个check检验,影响计算结果吗,不影响的话删除也可吧,不是所有的torch多的参数 都需要转写的

Copy link
Collaborator

@zhwesky2010 zhwesky2010 left a comment

Choose a reason for hiding this comment

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

文档修改后,读一读通顺吗?自己检查一下 不要直接按comment复制粘贴

@@ -6,7 +6,7 @@
setuptools.setup(*args, **attrs)
```

Paddle 无此 API,需要组合实现。该 API 一般情况下与 Paddle 无关,仅在 torch 分布式相关的深度学习用法里才需转写,用来构建一个包含自定义扩展(如 C++ )的 PyTorch 包。
Paddle 无此 API,需要组合实现。该 API 一般情况下与 Paddle 无关,用来构建一个包含自定义扩展(如 C++ )的 PyTorch 包。
Copy link
Collaborator

@zhwesky2010 zhwesky2010 Nov 18, 2024

Choose a reason for hiding this comment

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

这句话通顺吗?

Copy link
Collaborator

@zhwesky2010 zhwesky2010 left a comment

Choose a reason for hiding this comment

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

LGTM

@luotao1 luotao1 merged commit 5feeddb into PaddlePaddle:develop Nov 19, 2024
2 checks passed
WSBG pushed a commit to WSBG/docs that referenced this pull request Nov 27, 2024
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