mboillet commited on
Commit
39c4be0
·
verified ·
1 Parent(s): d64ad83

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -12
README.md CHANGED
@@ -25,39 +25,43 @@ This model performs Handwritten Text Recognition in Norwegian. It was developed
25
 
26
  ## Model description
27
 
28
- The model has been trained using the PyLaia library on the [NorHand v1](https://zenodo.org/record/6542056) document images.
29
 
30
  Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
31
 
32
- | split | N horizontal lines |
33
- | ----- | ------: |
34
  | train | 19,653 |
35
  | val | 2,286 |
36
  | test | 1,793 |
37
 
38
- An external 6-gram character language model can be used to improve recognition. The language model is trained on the text from the NorHand v1 training set.
39
-
40
  ## Evaluation results
41
 
42
  The model achieves the following results:
43
 
44
- | set | Language model | CER (%) | WER (%) | N lines |
45
  |:------|:---------------| ----------:| -------:|----------:|
46
  | test | no | 7.94 | 24.04 | 1,793 |
47
  | test | yes | 6.55 | 18.20 | 1,793 |
48
 
 
 
49
  ## How to use?
50
 
51
- Please refer to the [documentation](https://atr.pages.teklia.com/pylaia/).
52
 
53
  # Cite us!
54
 
55
  ```bibtex
56
- @inproceedings{pylaia-lib,
57
- author = "Tarride, Solène and Schneider, Yoann and Generali, Marie and Boillet, Melodie and Abadie, Bastien and Kermorvant, Christopher",
58
- title = "Improving Automatic Text Recognition with Language Models in the PyLaia Open-Source Library",
59
- booktitle = "Submitted at ICDAR2024",
60
- year = "2024"
 
 
 
 
61
  }
62
  ```
63
 
 
25
 
26
  ## Model description
27
 
28
+ The model has been trained using the PyLaia library on the [NorHand v1](https://zenodo.org/record/6542056) dataset.
29
 
30
  Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
31
 
32
+ | set | horizontal lines |
33
+ | :---- | ------: |
34
  | train | 19,653 |
35
  | val | 2,286 |
36
  | test | 1,793 |
37
 
 
 
38
  ## Evaluation results
39
 
40
  The model achieves the following results:
41
 
42
+ | set | Language model | CER (%) | WER (%) | lines |
43
  |:------|:---------------| ----------:| -------:|----------:|
44
  | test | no | 7.94 | 24.04 | 1,793 |
45
  | test | yes | 6.55 | 18.20 | 1,793 |
46
 
47
+ An external 6-gram character language model can be used to improve recognition. The language model is trained on the text from the NorHand v1 training set.
48
+
49
  ## How to use?
50
 
51
+ Please refer to the [PyLaia documentation](https://atr.pages.teklia.com/pylaia/usage/prediction/) to use this model.
52
 
53
  # Cite us!
54
 
55
  ```bibtex
56
+ @inproceedings{pylaia2024,
57
+ author = {Tarride, Solène and Schneider, Yoann and Generali-Lince, Marie and Boillet, Mélodie and Abadie, Bastien and Kermorvant, Christopher},
58
+ title = {{Improving Automatic Text Recognition with Language Models in the PyLaia Open-Source Library}},
59
+ booktitle = {Document Analysis and Recognition - ICDAR 2024},
60
+ year = {2024},
61
+ publisher = {Springer Nature Switzerland},
62
+ address = {Cham},
63
+ pages = {387--404},
64
+ isbn = {978-3-031-70549-6}
65
  }
66
  ```
67