Skip to content

PPstructureV3中,使用版面检测相关模型报错 #15847

Open
@free-all321

Description

@free-all321

🔎 Search before asking

  • I have searched the PaddleOCR Docs and found no similar bug report.
  • I have searched the PaddleOCR Issues and found no similar bug report.
  • I have searched the PaddleOCR Discussions and found no similar bug report.

🐛 Bug (问题描述)

使用PPstructureV3时,版面检测部分报错。
使用paddlepaddle3.0的docker镜像,在官网下载的的PP-DocLayout-L_infer模型。
使用V2版本中picodet_lcnet_x1_0_fgd_layout_infer模型则不会报错。

File "/home/paddle/PaddleOCR-main/ppstructure/layout/predict_layout.py", line 96, in call
self.predictor.run()
ValueError: In user code:

InvalidArgumentError: The input of Op(Conv) should be a 4-D or 5-D Tensor. But received: input's dimension is -1, input's shape is [].
  [Hint: Expected in_dims.size() == 4 || in_dims.size() == 5 == true, but received in_dims.size() == 4 || in_dims.size() == 5:0 != true:1.] (at ../paddle/phi/kernels/cpu/conv_util.h:130)
  [operator < pd_kernel.phi_kernel > error]

🏃‍♂️ Environment (运行环境)

paddlepaddle3.0的docker镜像
debian

🌰 Minimal Reproducible Example (最小可复现问题的Demo)

1、使用paddleOCR-main
2、使用PP-DocLayout-L_infer模型
3、python ppstructure/layout/predict_layout.py

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions