Rut5 Small Chitchat

🚀 Fine-tuned Russian Dialogue Model

This is a fine-tuned version of the cointegrated/rut5-small model on Russian dialogue data. It offers a balance of speed and size, suitable for chatbot fallback responses or style imitation.

🚀 Quick Start

This model is a fine-tuned variant of the cointegrated/rut5-small model, trained on certain Russian dialogue data. While it may not be highly intelligent or creative, its small size and fast processing speed make it useful as a fallback response generator for chatbots. Additionally, it can be further fine-tuned to mimic someone's style.

The model takes previous dialogue utterances separated by '\n\n' as input and generates the next utterance as output.

💻 Usage Examples

Basic Usage

# !pip install transformers sentencepiece
import torch
from transformers import T5ForConditionalGeneration, T5Tokenizer

tokenizer = T5Tokenizer.from_pretrained("cointegrated/rut5-small-chitchat")
model = T5ForConditionalGeneration.from_pretrained("cointegrated/rut5-small-chitchat")

text = 'Привет! Расскажи, как твои дела?'
inputs = tokenizer(text, return_tensors='pt')
with torch.no_grad():
    hypotheses = model.generate(
        **inputs, 
        do_sample=True, top_p=0.5, num_return_sequences=3, 
        repetition_penalty=2.5,
        max_length=32,
    )
for h in hypotheses:
    print(tokenizer.decode(h, skip_special_tokens=True))
# Как обычно.
# Сейчас - в порядке.
# Хорошо.
# Wall time: 363 ms 

📄 License

This project is licensed under the MIT license.