wjbmattingly commited on
Commit
4682866
1 Parent(s): 56a0241

added catmus citation

Browse files
Files changed (2) hide show
  1. README.md +44 -0
  2. send_image.py +2 -2
README.md CHANGED
@@ -10,3 +10,47 @@ license: cc-by-sa-4.0
10
 
11
  This is a simple [Kraken](https://kraken.re/main/index.html) FastAPI. It is designed to allow users to obtain line segmentation simply from a Kraken model. I have plans to expand this to include all other Kraken inputs including OCR.
12
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
10
 
11
  This is a simple [Kraken](https://kraken.re/main/index.html) FastAPI. It is designed to allow users to obtain line segmentation simply from a Kraken model. I have plans to expand this to include all other Kraken inputs including OCR.
12
 
13
+
14
+ # Models
15
+
16
+ ## Catmus Medieval
17
+
18
+ ```bibtex
19
+ @article{pinche_clérice_chagué_camps_vlachou-efstathiou_gille levenson_brisville-fertin_boschetti_fischer_gervers_et al._2024,
20
+ title={CATMuS Medieval},
21
+ DOI={10.5281/zenodo.12743230},
22
+ abstractNote={CATMuS (Consistent Approach to Transcribing ManuScript) Medieval is a Kraken HTR model trained on four different languages (in descending order of importance in the dataset: Old and Middle French, Latin, Spanish (and other languages of Spain), Italian) on strictly graphematic transcriptions. No abbreviations are resolved.
23
+
24
+ This model is the result of the collaboration from researchers from CREMMA, GalliCorpora, HTRomance and DEEDS projects. It follows the CREMMA Guidelines (Supplemented by the CREMMA Medii Aevi) and will be consolidated under the CATMuS Medieval Guidelines in an upcoming paper.
25
+
26
+ The model is trained with NFD Unicode normalization: each diacritic (including superscripts) are transcribed as their own characters, separately from the "main" character.
27
+
28
+ Data
29
+
30
+ See https://huggingface.co/datasets/CATMuS/medieval
31
+
32
+ All source datasets and papers are referenced in the related works section, all transcribers are mentioned in the collaborators section, all partner-project members are mentioned as authors.
33
+
34
+ Fundings
35
+
36
+
37
+
38
+ CREMMA, DIM MAP, Région Île-de-France
39
+
40
+ CremmaLab, DIM MAP, Région Île-de-France
41
+
42
+ GalliCorpora, Datalab, Bibliothèque nationale de France
43
+
44
+ HTRomance, Datalab, Bibliothèque nationale de France
45
+
46
+ Text as Image, Image as Text: Charter integrity and topic modelling, SSHRCC 1350911
47
+
48
+ Les Décades de Bersuire, première traduction française de l'Histoire romaine de Tite-Live – LiBer, ANR 21-CE27-0008
49
+
50
+ Projet Fabliaux, Biblissima+, ANR 21-ESRE-0005},
51
+ publisher={Zenodo},
52
+ author={Pinche, Ariane and Clérice, Thibault and Chagué, Alix and Camps, Jean-Baptiste and Vlachou-Efstathiou, Malamatenia and Gille Levenson, Matthias and Brisville-Fertin, Olivier and Boschetti, Federico and Fischer, Franz and Gervers, Michael and et al.},
53
+ year={2024},
54
+ month={Jul}
55
+ }
56
+ ```
send_image.py CHANGED
@@ -2,8 +2,8 @@ import requests
2
  import os
3
 
4
  # API endpoint
5
- # url = "https://wjbmattingly-kraken-api.hf.space/ocr"
6
- url = "http://127.0.0.1:8000/process_all"
7
 
8
  # Path to the image file
9
  image_path = os.path.join("data", "ms.jpg")
 
2
  import os
3
 
4
  # API endpoint
5
+ url = "https://wjbmattingly-kraken-api.hf.space/ocr"
6
+ # url = "http://127.0.0.1:8000/process_all"
7
 
8
  # Path to the image file
9
  image_path = os.path.join("data", "ms.jpg")