Rubert Base Cased Sentiment Rurewiews

🚀 RuBERT for Sentiment Analysis of Product Reviews

This project provides a model based on DeepPavlov/rubert-base-cased-conversational, trained on the RuReviews dataset for sentiment analysis of product reviews.

🚀 Quick Start

This is a DeepPavlov/rubert-base-cased-conversational model trained on RuReviews.

✨ Features

Labels

• 0: NEUTRAL
• 1: POSITIVE
• 2: NEGATIVE

💻 Usage Examples

Basic Usage

import torch
from transformers import AutoModelForSequenceClassification
from transformers import BertTokenizerFast

tokenizer = BertTokenizerFast.from_pretrained('blanchefort/rubert-base-cased-sentiment-rurewiews')
model = AutoModelForSequenceClassification.from_pretrained('blanchefort/rubert-base-cased-sentiment-rurewiews', return_dict=True)

@torch.no_grad()
def predict(text):
    inputs = tokenizer(text, max_length=512, padding=True, truncation=True, return_tensors='pt')
    outputs = model(**inputs)
    predicted = torch.nn.functional.softmax(outputs.logits, dim=1)
    predicted = torch.argmax(predicted, dim=1).numpy()
    return predicted

📚 Documentation

Dataset used for model training

RuReviews

RuReviews: An Automatically Annotated Sentiment Analysis Dataset for Product Reviews in Russian.