Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -38,7 +38,7 @@ We unified the label scheme into three sentiment classes:
|
|
| 38 |
## Dataset Structure
|
| 39 |
### Data Fields
|
| 40 |
- **text**: (string) Input sentence in Bengali-English code-mixed text
|
| 41 |
-
- **label**: (string) Sentiment label → `positive`, `negative`, `neutral`
|
| 42 |
|
| 43 |
### Splits
|
| 44 |
- **train**: 16,012 examples
|
|
|
|
| 38 |
## Dataset Structure
|
| 39 |
### Data Fields
|
| 40 |
- **text**: (string) Input sentence in Bengali-English code-mixed text
|
| 41 |
+
- **label**: (string) Sentiment label → `positive`, `negative`, `neutral`, `mixed`
|
| 42 |
|
| 43 |
### Splits
|
| 44 |
- **train**: 16,012 examples
|