|
|
--- |
|
|
tags: |
|
|
- text-classification |
|
|
library_name: transformers |
|
|
datasets: |
|
|
- I77/valid_invalid_questions |
|
|
language: |
|
|
- ru |
|
|
metrics: |
|
|
- accuracy |
|
|
- f1 |
|
|
base_model: |
|
|
- DeepPavlov/rubert-base-cased |
|
|
--- |
|
|
|
|
|
# RuBERT Model for Logical Question Classification |
|
|
|
|
|
This model classifies questions as logically correct or incorrect. |
|
|
|
|
|
1 - logically correct |
|
|
|
|
|
0 - logically incorrect |
|
|
|
|
|
# Metrics on validation |
|
|
|
|
|
Accuracy: 0.9500 |
|
|
Precision: 1.0000 |
|
|
Recall: 0.9000 |
|
|
F1-score: 0.9474 |
|
|
|
|
|
|
|
|
### Example inference |
|
|
|
|
|
```python |
|
|
from transformers import pipeline |
|
|
classifier = pipeline('text-classification', model='I77/question_classifier') |
|
|
print(classifier('Этот вопрос логичен?')) |
|
|
``` |
