|
--- |
|
language: ru |
|
license: apache-2.0 |
|
datasets: |
|
- IlyaGusev/gazeta |
|
--- |
|
|
|
# RuT5LargeSumGazeta |
|
|
|
## Model description |
|
|
|
This is the model for abstractive summarization for Russian based on ai-forever/ruT5-large. |
|
|
|
## Intended uses & limitations |
|
|
|
### How to use |
|
|
|
Here is how to use this model in PyTorch: |
|
|
|
```python |
|
from transformers import AutoTokenizer, T5ForConditionalGeneration |
|
|
|
model_name = "mlenjoyneer/rut5_large_sum_gazeta" |
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
model = T5ForConditionalGeneration.from_pretrained(model_name) |
|
|
|
article_text = "..." |
|
|
|
input_ids = tokenizer( |
|
[article_text], |
|
max_length=600, |
|
add_special_tokens=True, |
|
padding="max_length", |
|
truncation=True, |
|
return_tensors="pt" |
|
)["input_ids"] |
|
|
|
output_ids = model.generate( |
|
input_ids=input_ids, |
|
no_repeat_ngram_size=4 |
|
)[0] |
|
|
|
summary = tokenizer.decode(output_ids, skip_special_tokens=True) |
|
print(summary) |
|
``` |
|
|
|
## Training data |
|
|
|
- Dataset: [Gazeta](https://huggingface.co/datasets/IlyaGusev/gazeta) |
|
|
|
## Evaluation results |
|
|
|
| Model | R-1-f | R-2-f | R-L-f | chrF | BLEU | Avg char length | |
|
| -------------------------------------------- | ----- | ----- | ----- | ---- | ---- | --------------- | |
|
| IlyaGusev/mbart_ru_sum_gazeta | 28.7 | 11.1 | 24.4 | **37.3** | **9.4** | 373 | |
|
| IlyaGusev/rut5_base_sum_gazeta | 28.6 | 11.1 | 24.5 | 37.2 | **9.4** | 331 | |
|
| IlyaGusev/rugpt3medium_sum_gazeta | 24.1 | 6.5 | 19.8 | 32.1 | 3.6 | 242 | |
|
| rut5-large_sum_gazeta | **29.6** | **11.7** | **25.2** | **37.3** | **9.4** | 304 | |
|
|
