[xdoctest][task 184-185] reformat example code with google style in distributed/auto_parallel/static/*
#56666
Conversation
This file contains hidden or 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
megemini
reviewed
Aug 26, 2023
Comment on lines
+61
to
+68
| >>> import paddle | ||
| >>> import paddle.distributed as dist | ||
|
|
||
| paddle.enable_static() | ||
| >>> paddle.enable_static() | ||
|
|
||
| mesh = dist.DeviceMesh([[2, 4, 5], [0, 1, 3]]) | ||
| assert mesh.shape == [2, 3] | ||
| assert mesh.device_ids == [2, 4, 5, 0, 1, 3] | ||
| >>> mesh = dist.DeviceMesh([[2, 4, 5], [0, 1, 3]]) | ||
| >>> assert mesh.shape == [2, 3] | ||
| >>> assert mesh.device_ids == [2, 4, 5, 0, 1, 3] |
There was a problem hiding this comment.
distributed 相关的代码，先加上 # doctest: +REQUIRES(env:DISTRIBUTED) 吧 ～
这个 xxx_v2 感觉像是临时的 ... ...
|
@megemini 可以再review一下 |
megemini
reviewed
Aug 31, 2023
Comment on lines
357
to
365
| >>> # doctest: +REQUIRES(env:DISTRIBUTED) | ||
| >>> import numpy as np | ||
| >>> partition_tensor_list = [(np.array([[[1.11, 1.12]]]), [[0,1],[0,1],[0,2]])] | ||
| >>> tensor = np.array([[[1.13, 1.14]]]) | ||
| >>> partition_index = [[0,1],[0,1],[2,4]] | ||
|
|
||
| _merge_tensor(partition_tensor_list, tensor, partition_index) | ||
| # partition_tensor_list: [(np.array([[[1.11, 1.12, 1.13, 1.14]]]), [[0,1],[0,1],[0,4]])] | ||
| >>> _merge_tensor(partition_tensor_list, tensor, partition_index) | ||
| >>> print(partition_tensor_list) | ||
| [(np.array([[[1.11, 1.12, 1.13, 1.14]]]), [[0,1],[0,1],[0,4]])] |
There was a problem hiding this comment.
把 _merge_tensor 改为 merge 吧：
>>> import numpy as np
>>> import paddle
>>> from paddle.distributed.auto_parallel.static.converter import Converter
>>> partition_tensor_list = [(np.array([[[1.11, 1.12]]]), [[0,1],[0,1],[0,2]])]
>>> tensor = np.array([[[1.13, 1.14]]])
>>> partition_index = [[0,1],[0,1],[2,4]]
>>> complete_shape = [3, 2]
>>> Converter.merge(partition_tensor_list, tensor, partition_index, complete_shape)
>>> print(partition_tensor_list)
[(array([[[1.11, 1.12, 1.13, 1.14]]]), [[0, 1], [0, 1], [0, 4]])]
Comment on lines
423
to
434
| >>> # doctest: +REQUIRES(env:DISTRIBUTED) | ||
| >>> import numpy as np | ||
| >>> complete_tensor = np.array([[[1.11, 1.12, 1.13, 1.14, 1.15, 1.16]]]) | ||
| >>> rank = 2 | ||
| >>> complete_shape = [1, 1, 6] | ||
| >>> dims_mapping = [-1, -1, 0] | ||
| >>> process_shape = [3] | ||
| >>> process_group = [0, 1, 2] | ||
|
|
||
| >>> sliced_tensor_list = split(complete_tensor, [[], [], [2, 4]], 3) | ||
| >>> print(sliced_tensor_list) | ||
| [array([[[1.11, 1.12]]]), array([[[1.13, 1.14]]]), array([[[1.15, 1.16]]])] |
There was a problem hiding this comment.
split 用 Converter 静态调用：
>>> import numpy as np
>>> from paddle.distributed.auto_parallel.static.converter import Converter
>>> complete_tensor = np.array([[[1.11, 1.12, 1.13, 1.14, 1.15, 1.16]]])
>>> rank = 2
>>> complete_shape = [1, 1, 6]
>>> dims_mapping = [-1, -1, 0]
>>> process_shape = [3]
>>> process_group = [0, 1, 2]
>>> sliced_tensor_list = Converter.split(complete_tensor, [[], [], [2, 4]], 3)
>>> print(sliced_tensor_list)
[array([[[1.11, 1.12]]]), array([[[1.13, 1.14]]]), array([[[1.15, 1.16]]])]
Comment on lines
515
to
531
| >>> import numpy as np | ||
| >>> from paddle.distributed.auto_parallel.static.utils import _get_sliced_index | ||
| >>> complete_tensor = np.array([[[1.11, 1.12, 1.13, 1.14, 1.15, 1.16]]]) | ||
| >>> rank = 2 | ||
| >>> complete_shape = [1, 1, 6] | ||
| >>> dims_mapping = [-1, -1, 0] | ||
| >>> process_shape = [3] | ||
| >>> process_group = [0, 1, 2] | ||
|
|
||
| >>> slice_tensor = _slice_tensor(complete_tensor, [[], [], [2, 4]], 3) | ||
| >>> print(slice_tensor) | ||
| [array([[[1.11, 1.12]]]), array([[[1.13, 1.14]]]), array([[[1.15, 1.16]]])] | ||
|
|
||
| >>> index = _get_sliced_index(rank, complete_shape, dims_mapping, | ||
| ... process_shape, process_group) | ||
| >>> print(index) | ||
| 2 |
There was a problem hiding this comment.
这里的 _slice_tensor 没啥用，_get_sliced_index 也用静态调用吧：
>>> import numpy as np
>>> from paddle.distributed.auto_parallel.static.converter import Converter
>>> complete_tensor = np.array([[[1.11, 1.12, 1.13, 1.14, 1.15, 1.16]]])
>>> rank = 2
>>> complete_shape = [1, 1, 6]
>>> dims_mapping = [-1, -1, 0]
>>> process_shape = [3]
>>> process_group = [0, 1, 2]
>>> index = Converter._get_sliced_index(rank, complete_shape, dims_mapping,
... process_shape, process_group)
>>> print(index)
2|
啊 ～～～ cluster_v2.py 怎么给删了 ～～～ 🤣🤣🤣 我的意思是，那个改成那样就挺好（有 doctest 就行） ～ 里面的方法没法调用，可能是文件本身的问题，先不管了 ～ 是我的问题，以后应该说的再清楚一点 ～～～ 麻烦恢复回来吧，之前改的就挺好 ～ |
麻烦顺师傅了，我的问题，当时没看仔细 |
megemini
reviewed
Sep 1, 2023
| @@ -58,14 +58,15 @@ class DeviceMesh(core.DeviceMesh): | |||
| Examples: | |||
| .. code-block:: python | |||
|
|
|||
| import paddle | |||
| import paddle.distributed as dist | |||
| >>> # doctest: +SKIP('code is wrong') | |||
There was a problem hiding this comment.
# doctest: +SKIP('code is wrong') -> # doctest: +REQUIRES(env:DISTRIBUTED)
luotao1
approved these changes
Sep 5, 2023
BeingGod
pushed a commit
to BeingGod/Paddle
that referenced
this pull request
Sep 9, 2023
…distributed/auto_parallel/static/*` (PaddlePaddle#56666) * [Doctest]fix No.184,185, test=docs_preview * add env skip * fix @staticmethod * fix * add xdoctest for v2 * fix
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
Description
修改如下文件的示例代码，使其通过
xdoctest检查：python/paddle/distributed/auto_parallel/static/cluster_v2.pypython/paddle/distributed/auto_parallel/static/converter.py这个pr问题比较多，cluster_v2中的dist.DeviceMesh，converter的第2，3，5都有些函数没有找到对应的
预览：
Related links
@sunzhongkai588 @SigureMo @megemini