yschneider commited on
Commit
bc7807d
1 Parent(s): 17491f8

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. README.md +67 -3
  2. model +0 -0
  3. syms.txt +91 -0
  4. weights.ckpt +3 -0
README.md CHANGED
@@ -1,3 +1,67 @@
1
- ---
2
- license: cc0-1.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: PyLaia
3
+ license: mit
4
+ tags:
5
+ - PyLaia
6
+ - PyTorch
7
+ - atr
8
+ - htr
9
+ - ocr
10
+ - historical
11
+ - handwritten
12
+ metrics:
13
+ - CER
14
+ - WER
15
+ language:
16
+ - fr
17
+ datasets:
18
+ - Teklia/PELLET-Casimir-Marius-line
19
+ pipeline_tag: image-to-text
20
+ ---
21
+
22
+ # PyLaia - PELLET Casimir Marius - Line level
23
+
24
+ This model performs Handwritten Text Recognition in French.
25
+
26
+ ## Model description
27
+
28
+ The model has been trained using the PyLaia library on the [PELLET Casimir Marius - Line level](Teklia/PELLET-Casimir-Marius-line) dataset.
29
+
30
+ Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
31
+
32
+ | set | lines |
33
+ | ----- | ----: |
34
+ | train | 842 |
35
+ | val | 125 |
36
+ | test | 122 |
37
+
38
+ ## Evaluation results
39
+
40
+ The model achieves the following results:
41
+
42
+ | set | CER (%) | WER (%) | text_line |
43
+ | ----- | ------: | ------: | --------: |
44
+ | train | 0 | 0 | 842 |
45
+ | val | 0 | 0 | 125 |
46
+ | test | 0 | 0 | 122 |
47
+
48
+ ## How to use?
49
+
50
+ Please refer to the [PyLaia documentation](https://atr.pages.teklia.com/pylaia/usage/prediction/) to use this model.
51
+
52
+ ## Cite us!
53
+
54
+ -->
55
+
56
+ ```bibtex
57
+ @inproceedings{pylaia2024,
58
+ author = {Tarride, Solène and Schneider, Yoann and Generali-Lince, Marie and Boillet, Mélodie and Abadie, Bastien and Kermorvant, Christopher},
59
+ title = {{Improving Automatic Text Recognition with Language Models in the PyLaia Open-Source Library}},
60
+ booktitle = {Document Analysis and Recognition - ICDAR 2024},
61
+ year = {2024},
62
+ publisher = {Springer Nature Switzerland},
63
+ address = {Cham},
64
+ pages = {387--404},
65
+ isbn = {978-3-031-70549-6}
66
+ }
67
+ ```
model ADDED
Binary file (1.13 kB). View file
 
syms.txt ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <ctc> 0
2
+ ! 1
3
+ " 2
4
+ ' 3
5
+ ( 4
6
+ ) 5
7
+ , 6
8
+ - 7
9
+ . 8
10
+ / 9
11
+ 0 10
12
+ 1 11
13
+ 2 12
14
+ 3 13
15
+ 4 14
16
+ 5 15
17
+ 6 16
18
+ 7 17
19
+ 8 18
20
+ 9 19
21
+ : 20
22
+ ; 21
23
+ ? 22
24
+ A 23
25
+ B 24
26
+ C 25
27
+ D 26
28
+ E 27
29
+ F 28
30
+ G 29
31
+ H 30
32
+ I 31
33
+ J 32
34
+ K 33
35
+ L 34
36
+ M 35
37
+ N 36
38
+ O 37
39
+ P 38
40
+ Q 39
41
+ R 40
42
+ S 41
43
+ T 42
44
+ U 43
45
+ V 44
46
+ W 45
47
+ X 46
48
+ Y 47
49
+ Z 48
50
+ a 49
51
+ b 50
52
+ c 51
53
+ d 52
54
+ e 53
55
+ f 54
56
+ g 55
57
+ h 56
58
+ i 57
59
+ j 58
60
+ k 59
61
+ l 60
62
+ m 61
63
+ n 62
64
+ o 63
65
+ p 64
66
+ q 65
67
+ r 66
68
+ s 67
69
+ t 68
70
+ u 69
71
+ v 70
72
+ w 71
73
+ x 72
74
+ y 73
75
+ z 74
76
+ ° 75
77
+ À 76
78
+ à 77
79
+ â 78
80
+ ç 79
81
+ è 80
82
+ é 81
83
+ ê 82
84
+ ë 83
85
+ ô 84
86
+ ù 85
87
+ û 86
88
+ œ 87
89
+ ’ 88
90
+ <unk> 89
91
+ <space> 90
weights.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e49b4233c9bea4cfbe9a35e239e1c25421ceed9b194b07f6ec3d96e1b1a69fc
3
+ size 43033372