Skip to content

Commit

Permalink
docs: Update docs and fix markdown render error (PaddlePaddle#13678)
Browse files Browse the repository at this point in the history
  • Loading branch information
GreatV authored Aug 16, 2024
1 parent 08d35f6 commit fe564a6
Show file tree
Hide file tree
Showing 2 changed files with 6 additions and 5 deletions.
10 changes: 5 additions & 5 deletions docs/algorithm/table_recognition/algorithm_table_slanet.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,11 +8,11 @@ comments: true

该算法由来自北京交通大学机器学习与认识计算研究团队的ocr识别队研发,其在PaddleOCR算法模型挑战赛 - 赛题二:通用表格识别任务中排行榜荣获一等奖,排行榜精度相比PP-Structure表格识别模型提升0.8%,推理速度提升3倍。优化思路如下:

- 1. 改善推理过程,至EOS停止,速度提升3倍
- 2. 升级Backbone为LCNetV2(SSLD版本)
- 3. 行列特征增强模块
- 4. 提升分辨率488至512
- 5. 三阶段训练策略
1. 改善推理过程,至EOS停止,速度提升3倍
2. 升级Backbone为LCNetV2(SSLD版本)
3. 行列特征增强模块
4. 提升分辨率488至512
5. 三阶段训练策略

在PubTabNet表格识别公开数据集上,算法复现效果如下:

Expand Down
1 change: 1 addition & 0 deletions mkdocs.yml
Original file line number Diff line number Diff line change
Expand Up @@ -341,6 +341,7 @@ nav:
- SAR: algorithm/text_recognition/algorithm_rec_sar.md
- SEED: algorithm/text_recognition/algorithm_rec_seed.md
- SVTR: algorithm/text_recognition/algorithm_rec_svtr.md
- SVTRv2: algorithm/text_recognition/algorithm_rec_svtrv2.md
- ViTSTR: algorithm/text_recognition/algorithm_rec_vitstr.md
- ABINet: algorithm/text_recognition/algorithm_rec_abinet.md
- VisionLAN: algorithm/text_recognition/algorithm_rec_visionlan.md
Expand Down

0 comments on commit fe564a6

Please sign in to comment.