Datasets:
Commit
•
77b200a
1
Parent(s):
1d8bc3f
Update README.md
Browse files
README.md
CHANGED
|
@@ -33,3 +33,62 @@ configs:
|
|
| 33 |
- split: test
|
| 34 |
path: data/test-*
|
| 35 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
- split: test
|
| 34 |
path: data/test-*
|
| 35 |
---
|
| 36 |
+
|
| 37 |
+
task_categories:
|
| 38 |
+
- image-to-text
|
| 39 |
+
language:
|
| 40 |
+
- fr
|
| 41 |
+
tags:
|
| 42 |
+
- OCR
|
| 43 |
+
- NLP
|
| 44 |
+
- TAL
|
| 45 |
+
pretty_name: Split ICDAR2017 dataset
|
| 46 |
+
---
|
| 47 |
+
This dataset is a filtered version of the ICDAR2017 Competition on Handwritten Text Recognition, focusing on monograph texts written between 1800 and 1900. It consists of a total of **957 documents**, divided into training, validation, and testing sets, and is designed for post-correction of OCR (Optical Character Recognition) text.
|
| 48 |
+
|
| 49 |
+
- **Total Documents**: 957
|
| 50 |
+
- **Training Set**: 765
|
| 51 |
+
- **Validation Set**: 95
|
| 52 |
+
- **Test Set**: 97
|
| 53 |
+
|
| 54 |
+
## Purpose
|
| 55 |
+
The dataset aims to improve the accuracy of digitized texts by providing a reliable Gold Standard (GS) for comparison and correction, specifically addressing the challenges of older texts.
|
| 56 |
+
|
| 57 |
+
## Structure
|
| 58 |
+
The dataset is organized as follows:
|
| 59 |
+
|
| 60 |
+
```plaintext
|
| 61 |
+
dataset/
|
| 62 |
+
├── train/
|
| 63 |
+
│ ├── file1.txt
|
| 64 |
+
│ ├── file2.txt
|
| 65 |
+
│ └── ...
|
| 66 |
+
├── dev/
|
| 67 |
+
│ ├── file1.txt
|
| 68 |
+
│ ├── file2.txt
|
| 69 |
+
│ └── ...
|
| 70 |
+
├── test/
|
| 71 |
+
│ ├── file1.txt
|
| 72 |
+
│ ├── file2.txt
|
| 73 |
+
│ └── ...
|
| 74 |
+
└── metadata.csv # This file contains metadata for each txt file
|
| 75 |
+
```
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
- **Content** [#.txt]
|
| 79 |
+
- **1st line**: "[OCR_toInput] " => Raw OCRed text to be denoised.
|
| 80 |
+
- **2nd line**: "[OCR_aligned] " => Aligned OCRed text.
|
| 81 |
+
- **3rd line**: "[GS_aligned] " => Aligned Gold Standard.
|
| 82 |
+
|
| 83 |
+
The alignment was made at the character level using "@" symbols. "#" symbols correspond to the absence of GS either related to alignment uncertainities or related to unreadable characters in the source document.
|
| 84 |
+
For a better view of the alignment, make sure to disable the "word wrap" option in your text editor.
|
| 85 |
+
|
| 86 |
+
## Author Information
|
| 87 |
+
Prepared by **Mikhail Biriuchinskii**, an engineer in Natural Language Processing at Sorbonne University.
|
| 88 |
+
|
| 89 |
+
## Original Dataset Reference
|
| 90 |
+
For more information, visit the original dataset source: [ICDAR2017 Competition on Post-OCR Text Correction](http://l3i.univ-larochelle.fr/ICDAR2017PostOCR).
|
| 91 |
+
|
| 92 |
+
## Copyright
|
| 93 |
+
The original corpus is publicly accessible, and I do not hold any rights to this deployment of the corpus.
|
| 94 |
+
|