Commit ·
8c1050b
1
Parent(s): 9ebf49f
Update README.md
Browse files
README.md
CHANGED
|
@@ -9,4 +9,19 @@ widget:
|
|
| 9 |
example_title: "Example 2"
|
| 10 |
- text: "Dzbanek spadł na podłogę: trrrrach!"
|
| 11 |
example_title: "Example 3"
|
| 12 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
example_title: "Example 2"
|
| 10 |
- text: "Dzbanek spadł na podłogę: trrrrach!"
|
| 11 |
example_title: "Example 3"
|
| 12 |
+
---
|
| 13 |
+
---
|
| 14 |
+
# HerBERT Polish Onomatopoeia Recognition
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
This model is a finetuned HerBERT large model (cased) on a sample of 800 manually sentences containing onomatopoeia from the Mirosław Bańko's Współczesny polski onomatopeikon: ikoniczność w języku (Bańko, 2008).
|
| 18 |
+
The model allows to detect onomatopoeia in the form of interjections in any given text.
|
| 19 |
+
|
| 20 |
+
## Model Details
|
| 21 |
+
|
| 22 |
+
### Model Description
|
| 23 |
+
|
| 24 |
+
- **Developed by** Szymon Pindur, Jagiellonian University in Krakow
|
| 25 |
+
- **Model type** Token classification
|
| 26 |
+
- **Language(s) (NLP)** Polish
|
| 27 |
+
- **Finetuned from model [optional]** herbert-large-cased
|