[xdoctest][task 171-180] reformat example code with google style in audio/* #57111
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! |
whisky-12
changed the title
megemini
reviewed
Sep 8, 2023
python/paddle/amp/auto_cast.py
Outdated
| >>> data = paddle.to_tensor(data) | ||
| >>> with paddle.amp.amp_guard(): | ||
| ... conv = conv2d(data) | ||
| ... assert conv.dtype == paddle.float16 |
There was a problem hiding this comment.
这里在 aistudio 里面会报错，但是，如果不使用 conv2d 的 bias，就可以：
>>> import numpy as np
>>> import paddle
>>> paddle.seed(2023)
>>> data = np.random.uniform(-1, 1, [10, 3, 32, 32]).astype('float32')
>>> conv2d = paddle.nn.Conv2D(3, 2, 3, bias_attr=False)
>>> data = paddle.to_tensor(data)
>>> with paddle.amp.amp_guard():
... conv = conv2d(data)
... assert conv.dtype == paddle.float16或者需要先转换：
>>> import numpy as np
>>> import paddle
>>> paddle.seed(2023)
>>> data = np.random.uniform(-1, 1, [10, 3, 32, 32]).astype('float32')
>>> conv2d = paddle.nn.Conv2D(3, 2, 3)
>>> conv2d = paddle.amp.amp_decorate(models=conv2d, level='O2')
>>> data = paddle.to_tensor(data)
>>> with paddle.amp.amp_guard():
... conv = conv2d(data)
... assert conv.dtype == paddle.float16@SigureMo 为什么用了 bias，float16 就不能用了？？？
Comment on lines
162
to
164
| >>> paddle.device.set_device('gpu') | ||
|
|
||
| # required: gpu | ||
| paddle.set_device("gpu") | ||
| tensor = paddle.randn([512, 512, 512], "float") | ||
| del tensor | ||
| paddle.device.cuda.empty_cache() | ||
| >>> paddle.set_device("gpu") |
python/paddle/amp/debugging.py
Outdated
Comment on lines
92
to
99
| ... dtype = 'float32' | ||
| ... x = paddle.rand([10, 2, 2], dtype=dtype) | ||
| ... model = MyLayer(dtype) | ||
| ... x[0] = float(0) | ||
| ... loss = model(x) | ||
| ... adam = paddle.optimizer.Adam(parameters=model.parameters()) | ||
| ... loss.backward() | ||
| ... adam.step() |
There was a problem hiding this comment.
这部分缩进有问题，应该不是 MyLayer 的一部分
python/paddle/amp/debugging.py
Outdated
| ... # out_data = func(self, *modified_args, **kwargs) | ||
| ... # File "test.py", line 10, in forward | ||
| ... # return 1/x * self._w+ self._b | ||
| ... #RuntimeError: (PreconditionNotMet) There are NAN or INF (num_nan=0, num_inf=4, num_zero=0) in [device=gpu:0, op=divide, tensor=, dtype=fp32]. |
python/paddle/amp/debugging.py
Outdated
Comment on lines
483
to
489
| >>> # <------------------------------------------------------- op list --------------------------------------------------------> | ||
| >>> # <--------------- Op Name ---------------- | -- FP16 Calls --- | -- BF16 Calls --- | --- FP32 Calls--- | -- Other Calls --> | ||
| >>> # conv2d | 1 | 0 | 0 | 0 | ||
| >>> # elementwise_add | 1 | 0 | 0 | 0 | ||
| >>> # reshape2 | 1 | 0 | 0 | 0 | ||
| >>> # transfer_dtype | 0 | 0 | 3 | 0 | ||
| >>> # <----------------------------------------------------- op count: 4 ------------------------------------------------------> |
There was a problem hiding this comment.
amp 在 cpu 和 gpu 行为不同，CI 默认用 cpu，这里最好 set_device('gpu')，并且加 doctest requires～
另外，这里的输出，与我在 aistudio 上的也不一样，需要检查一下 ～
python/paddle/amp/debugging.py
Outdated
Comment on lines
510
to
528
| >>> import paddle | ||
|
|
||
| >>> conv = paddle.nn.Conv2D(3, 2, 3) | ||
| >>> paddle.seed(2023) | ||
| >>> x = paddle.rand([10, 3, 32, 32]) | ||
|
|
||
| >>> paddle.amp.debugging.enable_operator_stats_collection() | ||
| >>> # AMP list including conv2d, elementwise_add, reshape2, cast (transfer_dtype) | ||
| >>> with paddle.amp.auto_cast(enable=True, level='O2'): | ||
| ... out = conv(x) | ||
| >>> # Print to the standard output. | ||
| >>> paddle.amp.debugging.disable_operator_stats_collection() | ||
| >>> # <------------------------------------------------------- op list --------------------------------------------------------> | ||
| >>> # <--------------- Op Name ---------------- | -- FP16 Calls --- | -- BF16 Calls --- | --- FP32 Calls--- | -- Other Calls --> | ||
| >>> # conv2d | 1 | 0 | 0 | 0 | ||
| >>> # elementwise_add | 1 | 0 | 0 | 0 | ||
| >>> # reshape2 | 1 | 0 | 0 | 0 | ||
| >>> # transfer_dtype | 0 | 0 | 3 | 0 | ||
| >>> # <----------------------------------------------------- op count: 4 ------------------------------------------------------> |
python/paddle/amp/debugging.py
Outdated
| ... # AMP list including conv2d, elementwise_add, reshape2, cast (transfer_dtype) | ||
| ... with paddle.amp.auto_cast(enable=True, level='O2'): | ||
| ... out = conv(x) | ||
| >>> # Print to the standard output. |
python/paddle/amp/debugging.py
Outdated
| >>> from paddle.base import core | ||
| >>> try: | ||
| ... import xlsxwriter as xlw | ||
| >>> except ImportError: |
megemini
reviewed
Sep 9, 2023
python/paddle/amp/debugging.py
Outdated
| ... subprocess.check_call( | ||
| ... ['python', '-m', 'pip', 'install', 'xlsxwriter==3.0.9'] | ||
| ... ) | ||
| ... import xlsxwriter as xlw |
python/paddle/profiler/profiler.py
Outdated
Comment on lines
431
to
445
| >>> class RandomDataset(paddle.io.Dataset): | ||
| ... def __init__(self, num_samples): | ||
| ... self.num_samples = num_samples | ||
|
|
||
| ... def __getitem__(self, idx): | ||
| ... image = paddle.rand(shape=[100], dtype='float32') | ||
| ... label = paddle.randint(0, 10, shape=[1], dtype='int64') | ||
| ... return image, label | ||
|
|
||
| ... def __len__(self): | ||
| ... return self.num_samples | ||
|
|
||
| >>> class SimpleNet(paddle.nn.Layer): | ||
| ... def __init__(self): | ||
| ... super().__init__() | ||
| ... self.fc = paddle.nn.Linear(100, 10) | ||
|
|
||
| ... def forward(self, image, label=None): | ||
| ... return self.fc(image) |
There was a problem hiding this comment.
class 里面不要留空行 ～ 空行使 class 截断，导致运行检查错误。
python/paddle/sparse/binary.py
Outdated
Comment on lines
264
to
266
| [[ 0., -1., 0., 0.], | ||
| [ 0., 2., -6., 0.], | ||
| [ 6., 8., 4., 8.]] |
python/paddle/sparse/binary.py
Outdated
Comment on lines
317
to
319
| [[ 0., -1., 0., 4.], | ||
| [ 0., -2., 0., 0.], | ||
| [ 2., 2., -4., -8.]] |
python/paddle/sparse/binary.py
Outdated
Comment on lines
359
to
361
| [[ 0., 0., 0., -4.], | ||
| [ 0., 0., 9., 0.], | ||
| [ 8., 15., 0., 0.]] |
python/paddle/sparse/binary.py
Outdated
Comment on lines
404
to
406
| [[ nan , -inf. , nan , -1. ], | ||
| [ nan , 0. , 1. , nan ], | ||
| [ 2. , 1.66666663, 0. , 0. ]] |
python/paddle/sparse/binary.py
Outdated
Comment on lines
445
to
448
| >>> paddle.sparse.is_same_shape(x, y) | ||
| >>> #True | ||
| >>> paddle.sparse.is_same_shape(x, z) | ||
| >>> #False |
|
标题不对吧，不是 audio 吧 |
|
辛苦各位对PR问题的指摘，以上NG最近两天修复 ，以上 再次感谢 辛苦了！ |
whisky-12
force-pushed
the
xdoctest-171-180
branch
from
33e1d98 to
d34b4ea
Compare
|
@whisky-12 这个 PR 为什么 close 了呢？ |
|
喔，知道了，误 force push 了 |
|
PR 咋没了 我这边准备提交代码呢 |
|
因为你 force push 了无额外 commit 的分支，导致自动 close 非必要不要 force push |
|
哦 那个这个PR还能重新open嘛 |
|
|
上午那时 是想 将这条支线的代码和最新的develop保持一致 ，然后就在更新了下仓库 |
|
目前提交完代码 ，提示需要新建PR |
|
非空就可以 reopen 了 |
感谢 以后谨慎更新了 |
Frida-a
pushed a commit
to Frida-a/Paddle
that referenced
this pull request
Oct 14, 2023
…udio/* (PaddlePaddle#57111) * [Doctest]fix No.171-180, test=docs_preview * [Doctest]fix No.171-180, test=docs_preview * fix python/paddle/amp/auto_cast.py * fix python/paddle/amp/debugging.py * fix python/paddle/amp/grad_scaler.py * fix python/paddle/device/__init__.py * fix python/paddle/profiler/utils.py * fix python/paddle/sparse/binary.py * fix python/paddle/sparse/creation.py * fix ci error; * remove prompt before output * skip some output --------- Co-authored-by: SigureMo <[email protected]>
danleifeng
pushed a commit
to danleifeng/Paddle
that referenced
this pull request
Nov 14, 2023
…udio/* (PaddlePaddle#57111) * [Doctest]fix No.171-180, test=docs_preview * [Doctest]fix No.171-180, test=docs_preview * fix python/paddle/amp/auto_cast.py * fix python/paddle/amp/debugging.py * fix python/paddle/amp/grad_scaler.py * fix python/paddle/device/__init__.py * fix python/paddle/profiler/utils.py * fix python/paddle/sparse/binary.py * fix python/paddle/sparse/creation.py * fix ci error; * remove prompt before output * skip some output --------- Co-authored-by: SigureMo <[email protected]>
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
之前 #56159 提交代码时 只修改了对应行号出的代码 ，导致文件中其他代码的修改遗漏 ，本次PR进行补充
修改如下文件的示例代码，使其通过
xdoctest检查：python/paddle/amp/auto_cast.pypython/paddle/amp/debugging.pypython/paddle/amp/grad_scaler.pypython/paddle/device/__init__.pypython/paddle/device/cuda/__init__.pypython/paddle/profiler/profiler.pypython/paddle/profiler/utils.pypython/paddle/sparse/binary.pypython/paddle/sparse/creation.py预览：
Related links
@sunzhongkai588 @SigureMo @megemini