【Hackathon No.10】新增 LogNormal API #46426
Merged
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
|
你的PR提交成功，感谢你对开源项目的贡献! |
MayYouBeProsperous
changed the title
MayYouBeProsperous
changed the title
cxxly
reviewed
Sep 28, 2022
|
|
||
|
|
||
| class LogNormal(TransformedDistribution): | ||
| r"""The Normal distribution with location `loc` and `scale` parameters. |
cxxly
reviewed
Sep 28, 2022
| In the above equation: | ||
|
|
||
| * :math:`loc = \mu`: is the means of the underlying Normal distribution. | ||
| * :math:`scale = \sigma`: is the stddevs of the underlying Normal distribution. |
There was a problem hiding this comment.
这里是指LogNormal的基础分布
cxxly
reviewed
Sep 28, 2022
| Args: | ||
| loc(int|float|list|tuple|numpy.ndarray|Tensor): The means of normal distribution. | ||
| scale(int|float|list|tuple|numpy.ndarray|Tensor): The stddevs of normal distribution. | ||
| name(str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`. |
cxxly
reviewed
Sep 28, 2022
|
|
||
| """ | ||
| return self.sample(shape, seed) | ||
|
|
There was a problem hiding this comment.
rsample不能直接调用sample，sample不支持反向，需要通过标准正态分布重参数化实现。如果目前Paddle现有功能确实无法支持实现Normal sample的重参数化，可以暂时raise NotImplementedError
There was a problem hiding this comment.
新的commit补充了rsample的实现，但是我不确定实现是否正确，想请教一下，应该用什么方法验证rsample支持反向呢？
There was a problem hiding this comment.
param = paddle.rand(...)
param.stop_gradient = False
d = paddle.distribution.xxx(param = param)
y = d.rsample(...)
paddle.grad(y, param)可以新提交个PR进行验证，补充相应测试用例
cxxly
reviewed
Sep 28, 2022
| @@ -63,7 +63,7 @@ def __init__(self, base, transforms): | |||
| chain = transform.ChainTransform(transforms) | |||
| if len(base.batch_shape + base.event_shape) < chain._domain.event_rank: | |||
| raise ValueError( | |||
| f"'base' needs to have shape with size at least {chain._domain.event_rank}, bug got {len(base_shape)}." | |||
| f"'base' needs to have shape with size at least {chain._domain.event_rank}, bug got {len(base.batch_shape + base.event_shape)}." | |||
Ligoml
previously approved these changes
Oct 8, 2022
| Args: | ||
| loc(int|float|list|tuple|numpy.ndarray|Tensor): The means of the underlying Normal distribution. | ||
| scale(int|float|list|tuple|numpy.ndarray|Tensor): The stddevs of the underlying Normal distribution. | ||
| name(str, optional): Name for the operation (optional, default is None). For more information, please refer to :ref:`api_guide_Name`. |
There was a problem hiding this comment.
不用，name参数是为了追踪静态图下运行过程，每个方法也要处理，是一个通用逻辑，后面我会统一处理。此处可以先删除
There was a problem hiding this comment.
嗯嗯好的，那辛苦把这个参数去掉吧 @MayYouBeProsperous
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
Describe
新增 LogNormal API