Hugging Face's logo

Datasets:
yash-ingle
/
ILID_Indian_Language_Identification_Dataset

Tasks:
Text Classification
Modalities:
Text
Formats:
text
Languages:
Assamese
Bengali
Bodo (India)
Size:
100K - 1M
ArXiv:
Tags:
language-identification
indian-languages
Libraries:
Datasets
License:
Dataset card Data Studio Files
xet
Community
2
yash-ingle commited on
Commit
5212676
·
verified ·
1 Parent(s): 0fb0594

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +37 -36
README.md CHANGED
@@ -4,7 +4,7 @@ license: mit
4
 
5
  # 🗣️ IDLD23: Indian Dialect and Language Dataset (23 Languages)
6
 
7
- **Authors:** [Yash Ingle](mailto:yash.ingle003@gmail.com), [Pruthwik Mishra](mailto:pruthwikmishra@aid.svnit.ac.in)
8
  **Institute:** Sardar Vallabhbhai National Institute of Technology (SVNIT), Surat, India
9
 
10
  ---
@@ -16,46 +16,47 @@ The **IDLD23** (Indian Language Identification Dataset) benchmark contains **230
16
  ---
17
 
18
  ## 📊 Dataset Statistics
19
-
20
- | Language | Code | Train | Dev | Test | Total |
21
- |----------|------|-------|-----|------|-------|
22
- | Assamese | asm | 8000 | 1000 | 1000 | 10000 |
23
- | Bengali | ben | 8000 | 1000 | 1000 | 10000 |
24
- | Bodo | brx | 8000 | 1000 | 1000 | 10000 |
25
- | Dogri | doi | 8000 | 1000 | 1000 | 10000 |
26
- | Konkani | gom | 8000 | 1000 | 1000 | 10000 |
27
- | Gujarati | guj | 8000 | 1000 | 1000 | 10000 |
28
- | Hindi | hin | 8000 | 1000 | 1000 | 10000 |
29
- | Kannada | kan | 8000 | 1000 | 1000 | 10000 |
30
- | Kashmiri | kas | 8000 | 1000 | 1000 | 10000 |
31
- | Maithili | mai | 8000 | 1000 | 1000 | 10000 |
32
- | Malayalam | mal | 8000 | 1000 | 1000 | 10000 |
33
- | Marathi | mar | 8000 | 1000 | 1000 | 10000 |
34
- | Manipuri | mni | 8000 | 1000 | 1000 | 10000 |
35
- | Nepali | npi | 8000 | 1000 | 1000 | 10000 |
36
- | Odia | ory | 8000 | 1000 | 1000 | 10000 |
37
- | Punjabi | pan | 8000 | 1000 | 1000 | 10000 |
38
- | Sanskrit | san | 8000 | 1000 | 1000 | 10000 |
39
- | Santali | sat | 8000 | 1000 | 1000 | 10000 |
40
- | Sindhi | snd | 8000 | 1000 | 1000 | 10000 |
41
- | Tamil | tam | 8000 | 1000 | 1000 | 10000 |
42
- | Telugu | tel | 8000 | 1000 | 1000 | 10000 |
43
- | Urdu | urd | 8000 | 1000 | 1000 | 10000 |
44
- | English | eng | 8000 | 1000 | 1000 | 10000 |
45
- | **Total** | — | **184000** | **23000** | **23000** | **230000** |
 
46
 
47
  ---
 
48
 
49
- ## 📁 Files Provided
 
 
 
 
 
50
 
51
- - `idld23_train.csv`: 80% training split.
52
- - `idld23_dev.csv`: 10% development/validation set.
53
- - `idld23_test.csv`: 10% test set.
54
- - `idld23_full.csv`: All 230K samples.
55
- - `idld23_mixed.csv`: A random sample combining all languages.
56
- - `idld23_summary.csv`: Summary statistics or language distribution.
57
 
58
- ---
59
 
60
  ## 📌 Tasks
61
 
 
4
 
5
  # 🗣️ IDLD23: Indian Dialect and Language Dataset (23 Languages)
6
 
7
+ **Authors:** [Yash Ingle](mailto:yash.ingle003@gmail.com), [Dr.Pruthwik Mishra](mailto:pruthwikmishra@aid.svnit.ac.in)
8
  **Institute:** Sardar Vallabhbhai National Institute of Technology (SVNIT), Surat, India
9
 
10
  ---
 
16
  ---
17
 
18
  ## 📊 Dataset Statistics
19
+ | Language | Code | Train | Dev | Test | Total |
20
+ |--------------|-----------|-------|------|------|--------|
21
+ | Assamese | asm | 8000 | 1000 | 1000 | 10000 |
22
+ | Bengali | ben | 8000 | 1000 | 1000 | 10000 |
23
+ | Bodo | brx | 8000 | 1000 | 1000 | 10000 |
24
+ | Dogri | doi | 8000 | 1000 | 1000 | 10000 |
25
+ | Konkani | gom | 8000 | 1000 | 1000 | 10000 |
26
+ | Gujarati | guj | 8000 | 1000 | 1000 | 10000 |
27
+ | Hindi | hin | 8000 | 1000 | 1000 | 10000 |
28
+ | Kannada | kan | 8000 | 1000 | 1000 | 10000 |
29
+ | Kashmiri | kas | 8000 | 1000 | 1000 | 10000 |
30
+ | Maithili | mai | 8000 | 1000 | 1000 | 10000 |
31
+ | Malayalam | mal | 8000 | 1000 | 1000 | 10000 |
32
+ | Marathi | mar | 8000 | 1000 | 1000 | 10000 |
33
+ | Manipuri (Bengali) | mni_Beng | 8000 | 1000 | 1000 | 10000 |
34
+ | Manipuri (Meitei) | mni_Mtei | 8000 | 1000 | 1000 | 10000 |
35
+ | Nepali | npi | 8000 | 1000 | 1000 | 10000 |
36
+ | Odia | ory | 8000 | 1000 | 1000 | 10000 |
37
+ | Punjabi | pan | 8000 | 1000 | 1000 | 10000 |
38
+ | Sanskrit | san | 8000 | 1000 | 1000 | 10000 |
39
+ | Santali | sat | 8000 | 1000 | 1000 | 10000 |
40
+ | Sindhi (Arabic) | snd_Arab | 8000 | 1000 | 1000 | 10000 |
41
+ | Sindhi (Devanagari) | snd_Deva | 8000 | 1000 | 1000 | 10000 |
42
+ | Tamil | tam | 8000 | 1000 | 1000 | 10000 |
43
+ | Telugu | tel | 8000 | 1000 | 1000 | 10000 |
44
+ | Urdu | urd | 8000 | 1000 | 1000 | 10000 |
45
+ | English | eng | 8000 | 1000 | 1000 | 10000 |
46
+ | **Total** | — | **200000** | **25000** | **25000** | **250000** |
47
 
48
  ---
49
+ 📁 Files Provided
50
 
51
+ - shuffed_train_sentences: Training sentences (80% split – 200,000 samples)
52
+ - shuffed_train_labels: Corresponding labels for training sentences
53
+ - shuffed_dev_sentences: Validation (dev) sentences (10% split – 25,000 samples)
54
+ - shuffed_dev_labels: Corresponding labels for dev sentences
55
+ - shuffed_test_sentences: Test sentences (10% split – 25,000 samples)
56
+ - shuffed_test_labels: Corresponding labels for test sentences
57
 
58
+ Note: Filenames use 'shuffed' instead of 'shuffled'. Consider renaming for clarity.
 
 
 
 
 
59
 
 
60
 
61
  ## 📌 Tasks
62