[AMP OP&Test] adjust test_elementwise_sub's tolerance, max_relative_error of grad and #50953
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
atol/rtol of output to 1e-3
|
你的PR提交成功，感谢你对开源项目的贡献! |
|
❌ The PR is not created using PR's template. You can refer to this Demo. |
ZzSean
changed the title
ZzSean
reviewed
Apr 7, 2023
Comment on lines
393
to
394
| class TestElementwiseBF16OP_scalar(TestElementwiseOp): | ||
| def setUp(self): |
ZzSean
reviewed
Apr 11, 2023
| self.outputs = {'Out': convert_float_to_uint16(self.outputs['Out'])} | ||
|
|
||
| def test_check_grad_normal(self): | ||
| self.check_grad(['X', 'Y'], 'Out') |
There was a problem hiding this comment.
需要用check_grad_with_place,否则跑到CPU会出错，下同
| @@ -90,15 +133,62 @@ def if_enable_cinn(self): | |||
| self.enable_cinn = False | |||
|
|
|||
|
|
|||
| class TestElementwiseSubFP16OP_ZeroDim1(TestElementwiseOp): | |||
| or not core.is_bfloat16_supported(core.CUDAPlace(0)), | ||
| "core is not compiled with CUDA and do not support bfloat16", | ||
| ) | ||
| class TestElementwiseBF16OP_ZeroDim1(TestElementwiseOp): |
There was a problem hiding this comment.
直接继承TestElementwiseBF16OP，下同
| 'X': convert_float_to_uint16(self.inputs['X']), | ||
| 'Y': convert_float_to_uint16(self.inputs['Y']), | ||
| } | ||
| self.outputs = {'Out': convert_float_to_uint16(self.outputs['Out'])} |
There was a problem hiding this comment.
少了self.if_check_prim() self.if_enable_cinn()
ZzSean
reviewed
Apr 12, 2023
| @@ -66,7 +71,49 @@ def if_check_prim(self): | |||
| self.check_prim = True | |||
|
|
|||
| def if_enable_cinn(self): | |||
| pass | |||
| self.enable_cii = False | |||
lanxianghit
approved these changes
Apr 13, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
Others
PR changes
Others
Describe
Adjust the fp16 tolerance, max_relative_error of grad and atol/rtol of output to 1e-3