-
Notifications
You must be signed in to change notification settings - Fork 270
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
【Hackathon No.20】为 Paddle 新增 cummax API #39
Conversation
Thanks for your contribution! |
PR 格式检查通过,你的PR 将接受Paddle 专家以及开源社区的review,请及时关注PR 动态 |
fix the redundant file
测试考虑的case如下: | ||
|
||
- 动态图,静态图,与 PyTorch 的结果保持一致; | ||
- NaN 处理:关注异常值,边界情况的处理; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
请详细说明测试用例
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
已增补
你的PR有最新反馈,请及时修改。 |
# 三、业内方案调研 | ||
|
||
## PyTorch |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
可以提供更多框架(如 numpy / tf 等)方案的调研,以及目前 paddle 内是否可以通过组合方式实现。
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
谢谢,很有启发性
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
调研了NumPy,以组合方式实现(np.maximum.accumulate()
)。Paddle内没有提供accumulate
方法,并且考虑对齐 cumsum
、cumprod
,提供原生实现为好。
测试考虑的case如下: | ||
|
||
- 动态图,静态图,与 PyTorch 的结果保持一致; | ||
- NaN 处理:关注异常值,边界情况的处理; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这里的异常值处理设计方案具体是什么呢?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
已增补在测试用例说明,采取同NumPy一致
reply to how to handle NaN
fix dependency version add several test case description
LGTM |
No description provided.