---
language: ["ru", "en"]
tags:
- russian
license: mit
widget:
- text: "translate ru to en: Интересный момент. Модель не видела русских диалогов, но может их понимать"
---
This pruned model of mt5-base [google/mt5-base](https://huggingface.co/google/mt5-base) with only some Rusian and English embeddings left. 


The model has been fine-tuned for several tasks:
* translation (opus100 dataset)
* dialog (daily dialog dataset)


How to use:
```
# !pip install transformers sentencepiece

from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, T5Tokenizer
import torch

model_name = 'artemnech/enrut5-base'

model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)

def generate(text, **kwargs):
    model.eval()
    inputs = tokenizer(text, return_tensors='pt')
    with torch.no_grad():
        hypotheses = model.generate(**inputs,  **kwargs)
    return tokenizer.decode(hypotheses[0], skip_special_tokens=True)

print(generate('translate ru to en: Интересный момент. Модель не видела русских диалогов, но может их понимать', num_beams=4,))
# The Model didn't see Russian dialogues, but can understand them.

print(generate("translate en to ru: The Model didn't see Russian dialogues, but can understand them.", num_beams=4,))
# Модель не видела русских диалога, но может понимать их.

print(generate('dialog: user1>>: Hello', num_beams=2))
# Hi

print(generate('dialog: user1>>: Hello user2>>: Hi user1>>: Would you like to drink something?', num_beams=2))
# I'd like to drink a cup of coffee. 

#An interesting point. The model has not seen Russian dialogues, but can understand them

print(generate('dialog: user1>>: Привет'))
# Hi

print(generate('dialog: user1>>: Привет user2>>: Hi user1>>: Хочешь выпить что-нибудь?', num_beams=2))
# I'd like to have a cup of coffee. 
```