Add jacobian and hessian #53331
Merged
Add jacobian and hessian #53331
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
already. 2. support non-inplace math operation via magical method overwriting.
…ograd.functional.Jacobian/Hessian
2. increase TIMEOUT to 100
2. refine docstring
|
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. |
python/paddle/autograd/autograd.py
Outdated
| The ``xs`` tuples are identical in one-to-one correspondence. | ||
|
|
||
| - When ``batch_axis=None``, only 0-dimensional Tensor or 1-dimensional Tensor is | ||
| supported, assuming the shape of ``xs`` is ``[N, ]``, the shape of ``ys`` is |
There was a problem hiding this comment.
该段开头部分，需要when对齐，如
- When
batch_axis=None, only 0-dimensional Tensor or 1-dimensional Tensor is
supported, assuming the shape ofxsis[N, ], the shape ofysis
[M, ], then the output Jacobian matrix shape is[M, N].
否则会出现解析错误，下同
python/paddle/autograd/autograd.py
Outdated
| Returns: | ||
|
|
||
| Union[Tuple[Tuple[Jacobian, ...], ...], Tuple[Jacobian, ...], Jacobian]: Jacobian(s) of ys | ||
| deriveted from xs. |
python/paddle/autograd/autograd.py
Outdated
| ``(([M1, M1], [M1, M2]), ([M2, M1], [M2, M2]))`` | ||
|
|
||
| - When ``batch_axis=None``, only 0-dimensional Tensor or 1-dimensional Tensor is | ||
| supported, assuming that the shape of ``xs`` is ``[N, ]``, and the shape of ``ys`` is ``[ ]``(0-dimensional Tensor), the final output is a single Hessian matrix whose shape is ``[N, N]``. |
There was a problem hiding this comment.
开头和when对齐，而不是和 -对齐，下同
python/paddle/autograd/autograd.py
Outdated
| Returns: | ||
|
|
||
| Union[Tuple[Tuple[Hessian, ...], ...], Tuple[Hessian, ...], Hessian]: Hessian(s) of ys | ||
| deriveted from xs. |
XieYunshen
previously approved these changes
Apr 26, 2023
sunzhongkai588
previously approved these changes
Apr 26, 2023
python/paddle/autograd/autograd.py
Outdated
| ): | ||
| xs_grad = xs_grad[0] | ||
| else: | ||
| xs_grad = paddle.incubate.autograd.grad(ys, xs, v) |
There was a problem hiding this comment.
正式接口的代码不应依赖incubate里的代码，
假设incubate里的代码全部删除或者发生修改，正式接口的代码需要能正常运行。
python/paddle/autograd/autograd.py
Outdated
| return xs | ||
|
|
||
|
|
||
| def _grad(ys, xs, v=None): |
There was a problem hiding this comment.
这个函数名比较短，是否能在函数名称上体现一下跟grad的区别？
担心被认为是基础接口被其他代码调用
HydrogenSulfate
dismissed stale reviews from sunzhongkai588 and XieYunshen
via
5ee5a7d
kolinwei
approved these changes
Apr 27, 2023
XieYunshen
approved these changes
Apr 27, 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
New features
PR changes
APIs
Description
Pcard-66961
Add 2 high-order differential API:
paddle.autograd.jacobian与paddle.autograd.hessian，and related unitest file.Move paddle.incubate.autograd.Jacobian/Hessian to paddle.incubate.autograd.functional.Jacobian/Hessian