Skip to content

字符定位训练模型导出为推理模型后,精度发生变化 #16169

@afs1955

Description

@afs1955

🔎 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)

训练模型导出为推理模型后,精度发生了变化

image

上图为训练模型推理结果

image

推理模型预测结果1(导出为静态图推理模型):

导出推理命令:

python export_model.py -c /root/pyProjcet/PaddleOCR3.1/PaddleOCR/configs/det/PP-OCRv3/PP-OCRv3_mobile_det.yml -o Global.checkpoints=/root/pyProjcet/PaddleOCR3.1/PaddleOCR/myTest/project1/test/train/Train/best_accuracy.pdparams Global.save_inference_dir=/root/pyProjcet/PaddleOCR3.1/PaddleOCR/myTest/project1/test/train/Train Global.export_with_pir=False

image

推理模型预测结果2(导出为动态图推理模型):

导出推理命令:

python export_model.py -c /root/pyProjcet/PaddleOCR3.1/PaddleOCR/configs/det/PP-OCRv3/PP-OCRv3_mobile_det_copy.yml -o Global.pretrained_model=/root/pyProjcet/PaddleOCR3.1/PaddleOCR/myTest/project1/test/train/Train/best_accuracy.pdparams Global.save_inference_dir=/root/pyProjcet/PaddleOCR3.1/PaddleOCR/myTest/project1/test/train/Train

更详细的反馈情查看 #16154

这是我在Discussions反馈的问题

🏃‍♂️ Environment (运行环境)

OS WSL2-Ubuntu-22.04
Python 3.10.18
PaddleOCR 3.1.0
Paddlepaddle-gpu 3.1.0
install git
RAM 64.00GB
CPU I5 I2400F
CUDA 11.8

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

model.zip

提供了本次训练模型,与导出的推理模型,训练配置文件为config.yml

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