pasukka/auto-parts-term-classifier-v.2

Model Card for Model ID

Model Details

Model Description

• Language(s) (NLP): Russian
• License: apache-2.0
• Finetuned from model: ai-forever/ruRoberta-large

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification

model = AutoModelForSequenceClassification.from_pretrained("pasukka/auto-parts-term-classifier-v.2")
tokenizer = AutoTokenizer.from_pretrained('ai-forever/ruRoberta-large')

term = 'фильтр топливный'
outputs = model.forward(**tokenizer(term, return_tensors='pt').to(device='cuda'))
id = outputs.logits.argmax(dim=1)[0].item()

print(model.config.id2label[id])

Result:

фильтр

Metrics