remiai3 commited on
Commit
e2f280a
Β·
verified Β·
1 Parent(s): aad6b3d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +22 -18
README.md CHANGED
@@ -1,12 +1,31 @@
1
  ---
2
  license: cc-by-4.0
3
  language:
4
- - hi
5
- - ta
6
- - te
7
  - am
 
 
 
 
 
 
 
8
  size_categories:
9
  - 100K<n<1M
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
10
  ---
11
 
12
 
@@ -25,13 +44,10 @@ The dataset is provided in **two parallel formats**:
25
  - **Parquet version** (faster loading for large-scale pipelines)
26
 
27
  ⚠️ The images inside both folders are **identical** β€” only the label file format differs.
28
-
29
  So per language:
30
  - 10,000 unique images
31
  - 10,000 duplicate copies (same images, different label format)
32
 
33
- ---
34
-
35
  ## πŸ—‚ Dataset Structure
36
 
37
  tiny-captcha-library/
@@ -64,14 +80,10 @@ Each language folder includes:
64
  - `images.zip` β†’ CAPTCHA images
65
  - Label file mapping image filename β†’ correct 4-character text
66
 
67
- ---
68
-
69
  ## 🌐 Supported Scripts
70
 
71
  Amharic, Armenian, Cherokee, Chinese, Georgian, Hanunoo, Japanese, Kaithi, Khmer, Lao, Lisu, Miao, Modi, Myanmar, Osage, Sharada, Siddham, Soyombo, Tai Tham, Tai Viet, Takri, Thaana, Tirhuta
72
 
73
- ---
74
-
75
  ## πŸ–Ό Sample CAPTCHA Styles
76
 
77
  | Script | Example |
@@ -100,8 +112,6 @@ Amharic, Armenian, Cherokee, Chinese, Georgian, Hanunoo, Japanese, Kaithi, Khmer
100
  | Thaana | ![](./Thaana.png) |
101
  | Tirhuta | ![](./Tirhuta.png) |
102
 
103
- ---
104
-
105
  ## 🏷 Label Format
106
 
107
  ### CSV
@@ -113,8 +123,6 @@ Amharic, Armenian, Cherokee, Chinese, Georgian, Hanunoo, Japanese, Kaithi, Khmer
113
  ### Parquet
114
  Same structure as CSV but stored in columnar format.
115
 
116
- ---
117
-
118
  ## πŸ“Š Dataset Statistics
119
 
120
  | Metric | Value |
@@ -125,8 +133,6 @@ Same structure as CSV but stored in columnar format.
125
  | Images Including Duplicates | 460,000 |
126
  | Characters per CAPTCHA | 4 |
127
 
128
- ---
129
-
130
  ## 🎯 Intended Use
131
 
132
  This dataset is designed for:
@@ -138,8 +144,6 @@ This dataset is designed for:
138
 
139
  🚫 Not intended for bypassing real-world CAPTCHA security systems.
140
 
141
- ---
142
-
143
  ## βš™οΈ Example Loading
144
 
145
  ```python
 
1
  ---
2
  license: cc-by-4.0
3
  language:
 
 
 
4
  - am
5
+ - hy
6
+ - zh
7
+ - ka
8
+ - ja
9
+ - km
10
+ - lo
11
+ - my
12
  size_categories:
13
  - 100K<n<1M
14
+ tags:
15
+ - Cherokee
16
+ - Hanunoo
17
+ - Kaithi
18
+ - Lisu
19
+ - Miao
20
+ - Osage
21
+ - Sharada
22
+ - Siddham
23
+ - Soyombo
24
+ - TaiTham
25
+ - TaiViet
26
+ - Takri
27
+ - Thaana
28
+ - Tirhuta
29
  ---
30
 
31
 
 
44
  - **Parquet version** (faster loading for large-scale pipelines)
45
 
46
  ⚠️ The images inside both folders are **identical** β€” only the label file format differs.
 
47
  So per language:
48
  - 10,000 unique images
49
  - 10,000 duplicate copies (same images, different label format)
50
 
 
 
51
  ## πŸ—‚ Dataset Structure
52
 
53
  tiny-captcha-library/
 
80
  - `images.zip` β†’ CAPTCHA images
81
  - Label file mapping image filename β†’ correct 4-character text
82
 
 
 
83
  ## 🌐 Supported Scripts
84
 
85
  Amharic, Armenian, Cherokee, Chinese, Georgian, Hanunoo, Japanese, Kaithi, Khmer, Lao, Lisu, Miao, Modi, Myanmar, Osage, Sharada, Siddham, Soyombo, Tai Tham, Tai Viet, Takri, Thaana, Tirhuta
86
 
 
 
87
  ## πŸ–Ό Sample CAPTCHA Styles
88
 
89
  | Script | Example |
 
112
  | Thaana | ![](./Thaana.png) |
113
  | Tirhuta | ![](./Tirhuta.png) |
114
 
 
 
115
  ## 🏷 Label Format
116
 
117
  ### CSV
 
123
  ### Parquet
124
  Same structure as CSV but stored in columnar format.
125
 
 
 
126
  ## πŸ“Š Dataset Statistics
127
 
128
  | Metric | Value |
 
133
  | Images Including Duplicates | 460,000 |
134
  | Characters per CAPTCHA | 4 |
135
 
 
 
136
  ## 🎯 Intended Use
137
 
138
  This dataset is designed for:
 
144
 
145
  🚫 Not intended for bypassing real-world CAPTCHA security systems.
146
 
 
 
147
  ## βš™οΈ Example Loading
148
 
149
  ```python