fix the div 0 error of sequence_concat #49963

Liyulingyue · 2023-01-20T11:04:37Z

PR types

Bug fixes

PR changes

APIs

Describe

paddle-bot · 2023-01-20T11:04:43Z

你的PR提交成功，感谢你对开源项目的贡献!
请关注后续CI自动化测试结果，详情请参考Paddle-CI手册。
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

Liyulingyue · 2023-01-21T06:10:00Z

不知道为什么加入语句似乎没有触发执行，需要一些提示和帮助

SigureMo · 2023-01-31T21:50:45Z

不知道为什么加入语句似乎没有触发执行，需要一些提示和帮助

2023-01-20 20:12:37 ======================================================================
2023-01-20 20:12:37 ERROR: test_errors (test_sequence_concat.TestSequenceConcatOpError)
2023-01-20 20:12:37 ----------------------------------------------------------------------
2023-01-20 20:12:37 Traceback (most recent call last):
2023-01-20 20:12:37   File "/workspace/Paddle/build/python/paddle/fluid/tests/unittests/sequence/test_sequence_concat.py", line 140, in test_errors
2023-01-20 20:12:37     self.assertRaises(ValueError, test_0_shape)
2023-01-20 20:12:37   File "/opt/_internal/cpython-3.7.0/lib/python3.7/unittest/case.py", line 743, in assertRaises
2023-01-20 20:12:37     return context.handle('assertRaises', args, kwargs)
2023-01-20 20:12:37   File "/opt/_internal/cpython-3.7.0/lib/python3.7/unittest/case.py", line 178, in handle
2023-01-20 20:12:37     callable_obj(*args, **kwargs)
2023-01-20 20:12:37   File "/workspace/Paddle/build/python/paddle/fluid/tests/unittests/sequence/test_sequence_concat.py", line 135, in test_0_shape
2023-01-20 20:12:37     x4_data = fluid.layers.data(name="x4", shape=[0], dtype='float32')
2023-01-20 20:12:37 AttributeError: module 'paddle.fluid.layers' has no attribute 'data'
2023-01-20 20:12:37 
2023-01-20 20:12:37 ----------------------------------------------------------------------

根据报错可以发现 paddle.fluid.layers.data 这个 API 已经不存在了，这是在清理 fluid 过程中 #49301 清理掉的一个 API，可以 merge 一下 develop 并参考该 PR 修改 paddle.fluid.layers.data 为 paddle.static.data，再看看是否还会有问题～

wanghuancoder

LGTM

iclementine · 2023-02-03T09:48:41Z

python/paddle/fluid/layers/sequence_lod.py

+        if input_x.dim() == 1 and 0 in input_x.shape:
+            raise ValueError(
+                'the shape of element in input list should not be 0.'
+            )


Is this change intended to prevent division by zero at SequenceConcatOp's InferShape?

If so, why only raise Exception when an input has shape [0]. In my opinion, when an input has a leading zero in its shape, like [0, 2, 3], an exception should be raised.

got it, I will try to solve it

Is this change intended to prevent division by zero at SequenceConcatOp's InferShape?

If so, why only raise Exception when an input has shape [0]. In my opinion, when an input has a leading zero in its shape, like [0, 2, 3], an exception should be raised.

Hi, in the follow case, an exception have been writen.

If I simplely require that 'the input size must not be 0', some collisions may occur.

So, could you tell me where should I modify or what should I do?

import paddle paddle.enable_static() x = paddle.static.data(name='x', shape=[1,1], dtype='float32', lod_level=1) y = paddle.static.data(name='y', shape=[1,0], dtype='float32', lod_level=1) out = paddle.static.nn.sequence_concat(input=[x, y])

ValueError: Variable 'y' has been created before. The previous shape is (1, 0), the new shape is (0, 1). They are not matched.

If I simplely require that 'the input size must not be 0', some collisions may occur.

emmmm，就这个问题而言，因为是 InferShape 里出现了除第一个维度的需求，'the input size must not be 0' 的检查是不是不太合适？when an input has a leading zero in its shape 对此的检查就刚刚好

Hi, in the follow case, an exception have been writen.

ValueError: Variable 'y' has been created before. The previous shape is (1, 0), the new shape is (0, 1). They are not matched.

这个错误与本问题无关啊，只要你重新使用相同 name 定义了不同 shape 的 Variable 就会出现此错误，这里上文应该有一个定义过的 shape=(1, 0) 的 y，删掉或者将这个重命名即可（Jupyter 直接 Restart kernel）

So, could you tell me where should I modify or what should I do?

直接在 Python 端修改：

- if input_x.dim() == 1 and 0 in input_x.shape: # 仅仅要求 shape != [0] + if input_x.shape[0] == 0: # 要求 shape != [0, ...]

或者直接在 C++ 端增加一样的检查，比如这里 71 行后就很合适，个人觉得直接 C++ 端更好一些？

Paddle/paddle/fluid/operators/sequence_ops/sequence_concat_op.cc

Lines 71 to 86 in 3e5a6df

batch_size += x_dim[0];

if (feature_size == 0) {

feature_size = phi::product(x_dim) / x_dim[0];

} else {

PADDLE_ENFORCE_EQ(

feature_size,

phi::product(x_dim) / x_dim[0],

platform::errors::InvalidArgument(

"Each input of SequenceConcatOp inputs must have same feature "

"size, But "

"the feature size we received is %d, the feature size of 1st "

"input is %d",

feature_size,

phi::product(x_dim) / x_dim[0]));

}

}

If I simplely require that 'the input size must not be 0', some collisions may occur.

emmmm，就这个问题而言，因为是 InferShape 里出现了除第一个维度的需求，'the input size must not be 0' 的检查是不是不太合适？when an input has a leading zero in its shape 对此的检查就刚刚好

Hi, in the follow case, an exception have been writen.

ValueError: Variable 'y' has been created before. The previous shape is (1, 0), the new shape is (0, 1). They are not matched.

这个错误与本问题无关啊，只要你重新使用相同 name 定义了不同 shape 的 Variable 就会出现此错误，这里上文应该有一个定义过的 shape=(1, 0) 的 y，删掉或者将这个重命名即可（Jupyter 直接 Restart kernel）

So, could you tell me where should I modify or what should I do?

直接在 Python 端修改：

- if input_x.dim() == 1 and 0 in input_x.shape: # 仅仅要求 shape != [0] + if input_x.shape[0] == 0: # 要求 shape != [0, ...]

或者直接在 C++ 端增加一样的检查，比如这里 71 行后就很合适，个人觉得直接 C++ 端更好一些？

Paddle/paddle/fluid/operators/sequence_ops/sequence_concat_op.cc

Lines 71 to 86 in 3e5a6df

batch_size += x_dim[0];

if (feature_size == 0) {

feature_size = phi::product(x_dim) / x_dim[0];

} else {

PADDLE_ENFORCE_EQ(

feature_size,

phi::product(x_dim) / x_dim[0],

platform::errors::InvalidArgument(

"Each input of SequenceConcatOp inputs must have same feature "

"size, But "

"the feature size we received is %d, the feature size of 1st "

"input is %d",

feature_size,

phi::product(x_dim) / x_dim[0]));

}

}

更新一下报错信息~

ValueError: (InvalidArgument) Each input of SequenceConcatOp inputs must have same feature size, But the feature size we received is 1, the feature size of 1st input is 0 [Hint: Expected feature_size == phi::product(x_dim) / x_dim[0], but received feature_size:1 != phi::product(x_dim) / x_dim[0]:0.] (at /paddle/paddle/fluid/operators/sequence_ops/sequence_concat_op.cc:84) [operator < sequence_concat > error]

我也是觉得在 c++ 端添加 x_dim[0] 不能为 0 的检查会更好一些。

P.S. 由于现在 paddle 对 zero-sized tensor 支持不够好，所以目前添加这样的检查（无论是在 c++ 端还是在 Python 端）我觉得都 ok.

不过其实为了计算 feature_size 大可以用 product(x_shape[1: ]) 的方式来计算而不必用 x_numel / x_shape[0] 的方式计算. （P.S. paddle 的 lod tensor （多层不定长序列）的第 0 维表示递归折叠后的长度，而后面的可以称为 feature_shape，表示序列中的基础元素的形状）

iclementine

LGTM

fix the div 0 error of sequence_concat

004c1b7

paddle-bot bot added contributor External developers status: proposed labels Jan 20, 2023

bug fix

499b526

luotao1 assigned luotao1 and wanghuancoder Jan 20, 2023

luotao1 mentioned this pull request Jan 20, 2023

good first issue lists #49927

Closed

Liyulingyue and others added 2 commits February 1, 2023 19:15

Merge branch 'PaddlePaddle:develop' into div0_4

a180f53

replace fluid with static

a9a83aa

wanghuancoder previously approved these changes Feb 3, 2023

View reviewed changes

luotao1 assigned iclementine Feb 3, 2023

iclementine reviewed Feb 3, 2023

View reviewed changes

fixed

23df7e2

Liyulingyue dismissed wanghuancoder’s stale review via 23df7e2 February 5, 2023 22:15

luotao1 closed this Feb 6, 2023

luotao1 reopened this Feb 6, 2023

PaddlePaddle locked and limited conversation to collaborators Feb 6, 2023

PaddlePaddle unlocked this conversation Feb 6, 2023

Liyulingyue and others added 3 commits February 6, 2023 20:13

fix error

70f5049

Merge branch 'PaddlePaddle:develop' into div0_4

976f983

Update test_sequence_concat.py

257c6b4

Liyulingyue requested review from iclementine, SigureMo and wanghuancoder and removed request for iclementine, SigureMo and wanghuancoder February 7, 2023 05:16

Liyulingyue requested a review from iclementine February 7, 2023 05:16

iclementine approved these changes Feb 7, 2023

View reviewed changes

iclementine merged commit 84fe2de into PaddlePaddle:develop Feb 7, 2023

Liyulingyue deleted the div0_4 branch February 21, 2023 12:49

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix the div 0 error of sequence_concat #49963

fix the div 0 error of sequence_concat #49963

Liyulingyue commented Jan 20, 2023

paddle-bot bot commented Jan 20, 2023

Liyulingyue commented Jan 21, 2023 •

edited

Loading

SigureMo commented Jan 31, 2023

wanghuancoder left a comment

iclementine Feb 3, 2023 •

edited

Loading

Liyulingyue Feb 4, 2023

Liyulingyue Feb 5, 2023

SigureMo Feb 5, 2023

Liyulingyue Feb 5, 2023

iclementine Feb 6, 2023

iclementine left a comment

	batch_size += x_dim[0];
	if (feature_size == 0) {
	feature_size = phi::product(x_dim) / x_dim[0];
	} else {
	PADDLE_ENFORCE_EQ(
	feature_size,
	phi::product(x_dim) / x_dim[0],
	platform::errors::InvalidArgument(
	"Each input of SequenceConcatOp inputs must have same feature "
	"size, But "
	"the feature size we received is %d, the feature size of 1st "
	"input is %d",
	feature_size,
	phi::product(x_dim) / x_dim[0]));
	}
	}

fix the div 0 error of sequence_concat #49963

fix the div 0 error of sequence_concat #49963

Conversation

Liyulingyue commented Jan 20, 2023

PR types

PR changes

Describe

paddle-bot bot commented Jan 20, 2023

Liyulingyue commented Jan 21, 2023 • edited Loading

SigureMo commented Jan 31, 2023

wanghuancoder left a comment

Choose a reason for hiding this comment

iclementine Feb 3, 2023 • edited Loading

Choose a reason for hiding this comment

Liyulingyue Feb 4, 2023

Choose a reason for hiding this comment

Liyulingyue Feb 5, 2023

Choose a reason for hiding this comment

SigureMo Feb 5, 2023

Choose a reason for hiding this comment

Liyulingyue Feb 5, 2023

Choose a reason for hiding this comment

iclementine Feb 6, 2023

Choose a reason for hiding this comment

iclementine left a comment

Choose a reason for hiding this comment

Liyulingyue commented Jan 21, 2023 •

edited

Loading

iclementine Feb 3, 2023 •

edited

Loading