[co63oc]

PR Category

Execute Infrastructure

PR Types

Bug fixes

Description

paddle.nn.functional.conv1d
paddle.nn.functional.conv2d
paddle.nn.functional.conv3d

Infermeta 去掉维度为0检查，如果filter_dims[1] 为0，结果对应维度设置为0，原有代码是设置filter_dims[0]，符号推导同样修改。
修改kernel，只用判断input 维度为0
torch不支持单独filter设置为0维度

kernel 使用到depthwise_conv 和 conv kernel， 修改 depthwise_conv 和 conv kernel

正向测试返回填充为0，需要修改tester/accuracy.py输出结果， PaddleAPITest 对输入增加了pad

python engine.py --accuracy=True --api_config='paddle.nn.functional.conv3d(Tensor([4, 3, 8, 0, 8],"float32"), Tensor([5, 3, 3, 3, 3],"float32"), None, padding=list[list[0,0,],list[0,0,],list[1,1,],list[2,2,],list[2,2,],], stride=1, dilation=1, groups=1, data_format="NCDHW", )'

反向测试filter为填充0

import torch
x=torch.randn([0,1,1])
y=torch.randn([1,1,1])
x.requires_grad=True
y.requires_grad=True
z=torch.nn.functional.conv1d(x, y, groups=1)
z.sum().backward()
y.grad

tensor([[[0.]]])

PaddleAPITest 测试
GPU测试为torch error

CPU测试为torch error

[paddle-bot]

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.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 76.92308% with 3 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@3efb8db). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
...interface/infer_symbolic_shape/binary_infer_sym.cc	66.66%	2 Missing ⚠️
paddle/phi/infermeta/binary.cc	85.71%	1 Missing ⚠️

❌ Your patch status has failed because the patch coverage (76.92%) is below the target coverage (90.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             develop   #73691   +/-   ##
==========================================
  Coverage           ?   76.92%           
==========================================
  Files              ?        2           
  Lines              ?       13           
  Branches           ?        0           
==========================================
  Hits               ?       10           
  Misses             ?        3           
  Partials           ?        0           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[co63oc]

/re-run all-failed

[DanielSun11]

LGTM

[DanielSun11]

涉及到符号推导的改动请 @luotao1 协调云飞看一下