yschneider starride-teklia commited on
Commit
bbcb824
1 Parent(s): 11a9fac

Add model trained on Esposalles (#1)

Browse files

- Add model trained on Esposalles (5c5871e541e6b39dad54d1a97fc1f58cdead55ce)


Co-authored-by: Solène Tarride <[email protected]>

Files changed (7) hide show
  1. README.md +33 -0
  2. language_model.arpa.gz +3 -0
  3. lexicon.txt +63 -0
  4. model +0 -0
  5. syms.txt +63 -0
  6. tokens.txt +63 -0
  7. weights.ckpt +3 -0
README.md CHANGED
@@ -1,3 +1,36 @@
1
  ---
 
2
  license: mit
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ library_name: PyLaia
3
  license: mit
4
+ tags:
5
+ - PyLaia
6
+ - PyTorch
7
+ - Handwritten text recognition
8
+ metrics:
9
+ - CER
10
+ - WER
11
+ language:
12
+ - ca
13
  ---
14
+
15
+ # Catalan handwritten text recognition
16
+
17
+ This model performs Handwritten Text Recognition in Catalan (19th century).
18
+
19
+ ## Model description
20
+
21
+ The model has been trained using the PyLaia library on the [Esposalles](https://rrc.cvc.uab.es/?ch=10&com=introduction/) dataset.
22
+ Training images were resized with a fixed height of 128 pixels, keeping the original aspect ratio.
23
+
24
+
25
+ ## Evaluation results
26
+
27
+ The model achieves the following results:
28
+
29
+ | set | Language model | CER (%) | WER (%) | N lines |
30
+ |:------|:---------------|:----------:|:-------:|----------:|
31
+ | test | no | 0.76 | 2.62 | 757 |
32
+ | test | yes | 1.04 | 3.38 | 757 |
33
+
34
+ ## How to use
35
+
36
+ Please refer to the [documentation](https://atr.pages.teklia.com/pylaia/).
language_model.arpa.gz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:441de98cf792820f17109412e78dc4f956d72163198190991ee75faa65fd06d4
3
+ size 1574104
lexicon.txt ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <ctc> <ctc>
2
+ # #
3
+ 0 0
4
+ 1 1
5
+ 2 2
6
+ 3 3
7
+ 4 4
8
+ 5 5
9
+ 6 6
10
+ 7 7
11
+ 8 8
12
+ 9 9
13
+ = =
14
+ A A
15
+ B B
16
+ C C
17
+ D D
18
+ E E
19
+ F F
20
+ G G
21
+ H H
22
+ I I
23
+ J J
24
+ L L
25
+ M M
26
+ N N
27
+ O O
28
+ P P
29
+ Q Q
30
+ R R
31
+ S S
32
+ T T
33
+ U U
34
+ V V
35
+ X X
36
+ Y Y
37
+ a a
38
+ b b
39
+ c c
40
+ d d
41
+ e e
42
+ f f
43
+ g g
44
+ h h
45
+ i i
46
+ j j
47
+ l l
48
+ m m
49
+ n n
50
+ o o
51
+ p p
52
+ q q
53
+ r r
54
+ s s
55
+ t t
56
+ u u
57
+ v v
58
+ x x
59
+ y y
60
+ z z
61
+ ç ç
62
+ <unk> <unk>
63
+ <space> <space>
model ADDED
Binary file (1.52 kB). View file
 
syms.txt ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <ctc> 0
2
+ # 1
3
+ 0 2
4
+ 1 3
5
+ 2 4
6
+ 3 5
7
+ 4 6
8
+ 5 7
9
+ 6 8
10
+ 7 9
11
+ 8 10
12
+ 9 11
13
+ = 12
14
+ A 13
15
+ B 14
16
+ C 15
17
+ D 16
18
+ E 17
19
+ F 18
20
+ G 19
21
+ H 20
22
+ I 21
23
+ J 22
24
+ L 23
25
+ M 24
26
+ N 25
27
+ O 26
28
+ P 27
29
+ Q 28
30
+ R 29
31
+ S 30
32
+ T 31
33
+ U 32
34
+ V 33
35
+ X 34
36
+ Y 35
37
+ a 36
38
+ b 37
39
+ c 38
40
+ d 39
41
+ e 40
42
+ f 41
43
+ g 42
44
+ h 43
45
+ i 44
46
+ j 45
47
+ l 46
48
+ m 47
49
+ n 48
50
+ o 49
51
+ p 50
52
+ q 51
53
+ r 52
54
+ s 53
55
+ t 54
56
+ u 55
57
+ v 56
58
+ x 57
59
+ y 58
60
+ z 59
61
+ ç 60
62
+ <unk> 61
63
+ <space> 62
tokens.txt ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <ctc>
2
+ #
3
+ 0
4
+ 1
5
+ 2
6
+ 3
7
+ 4
8
+ 5
9
+ 6
10
+ 7
11
+ 8
12
+ 9
13
+ =
14
+ A
15
+ B
16
+ C
17
+ D
18
+ E
19
+ F
20
+ G
21
+ H
22
+ I
23
+ J
24
+ L
25
+ M
26
+ N
27
+ O
28
+ P
29
+ Q
30
+ R
31
+ S
32
+ T
33
+ U
34
+ V
35
+ X
36
+ Y
37
+ a
38
+ b
39
+ c
40
+ d
41
+ e
42
+ f
43
+ g
44
+ h
45
+ i
46
+ j
47
+ l
48
+ m
49
+ n
50
+ o
51
+ p
52
+ q
53
+ r
54
+ s
55
+ t
56
+ u
57
+ v
58
+ x
59
+ y
60
+ z
61
+ ç
62
+ <unk>
63
+ <space>
weights.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2682fa6a47fc5c0bae6c7add075dfec1a42f8b502311df5defb5aa0b3c4f1b4d
3
+ size 42598044