Banking77 Intent Classifier

Fine-tuned DistilBERT for classifying customer support messages into 77 fine-grained banking intents.

Model Details

• Base model: distilbert-base-uncased
• Training data: PolyAI/banking77
• Classes: 77 banking intents
• Parameters: 67M
• Training: 3 epochs, learning rate 2e-5, batch size 32

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

tokenizer = AutoTokenizer.from_pretrained("Sylvianjoki/banking77-distilbert")
model = AutoModelForSequenceClassification.from_pretrained("Sylvianjoki/banking77-distilbert")

text = "I lost my card yesterday"
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=64)

with torch.no_grad():
    logits = model(**inputs).logits

predicted_class_id = logits.argmax().item()
print(predicted_class_id)