-
Notifications
You must be signed in to change notification settings - Fork 5.8k
[XPU] add depthwise_conv2d backward unit test #67777
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
Conversation
你的PR提交成功,感谢你对开源项目的贡献! |
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.
LGTM
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.
lgtm
@@ -130,6 +131,21 @@ def init_paddings(self): | |||
self.pad = [0, 1, 0, 2] | |||
self.padding_algorithm = "EXPLICIT" | |||
|
|||
def test_check_grad_no_input(self): |
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.
这些check_grad
是不是应该加在基类里面呢?如果加在子类里面,为啥只加了这一个子类,这样的话其它的子类感觉还是不会跑反向的吧。
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.
是可以加到基类里。
但这里depthwise_conv2d
有两个测试类(用_v2区分),对比了一下其实区别不大,而且每个测试类里不同测试子类的区别在于input_size
。为了节省测试时间和代码简化,只选择了一种测试子类添加反向。
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.
本来要测试的,就是在不同形状下面的梯度都能算对才行呀。其它算子的测试case也是,不同形状下面都要能跑过才行。
这个修改可以先合,后续再改一把吧?
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.
好,后续再改
PR Category
Custom Device
PR Types
Improvements
Description
add depthwise_conv2d backward unit test