Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,21 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- ta
|
| 5 |
+
pretty_name: solvari
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
### Data Fields
|
| 9 |
+
|
| 10 |
+
Each Parquet file contains:
|
| 11 |
+
- `text`: Raw Tamil text
|
| 12 |
+
- `words`: Word count
|
| 13 |
+
|
| 14 |
+
### Technical Specifications
|
| 15 |
+
|
| 16 |
+
- Format: Parquet
|
| 17 |
+
- Encoding: UTF-8
|
| 18 |
+
- Text Filter: Tamil (Unicode range: \u0b80-\u0bff)
|
| 19 |
+
- Tokenizer: AI4Bharat/indic-bert
|
| 20 |
+
|
| 21 |
+
Total Tokens - **93,877,412**
|