File size: 844 Bytes
39bfe46 763d29d 39bfe46 ae76615 39bfe46 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 |
---
license: openrail++
language:
- ru
tags:
- text-generation-inference
datasets:
- s-nlp/ru_paradetox
---
This is the detoxification baseline model trained on the [train](https://github.com/skoltech-nlp/russe_detox_2022/blob/main/data/input/train.tsv) part of "RUSSE 2022: Russian Text Detoxification Based on Parallel Corpora" competition. The source sentences are Russian toxic messages from Odnoklassniki, Pikabu, and Twitter platforms. The base model is [ruT5](https://huggingface.co/sberbank-ai/ruT5-base) provided from Sber.
**How to use**
```python
from transformers import T5ForConditionalGeneration, AutoTokenizer
base_model_name = 'sberbank-ai/ruT5-base'
model_name = 'SkolkovoInstitute/ruT5-base-detox'
tokenizer = AutoTokenizer.from_pretrained(base_model_name)
model = T5ForConditionalGeneration.from_pretrained(model_name)
``` |