Branch Switch Classification Model

This model classifies whether a user wants to switch hospital branches or is asking for general information.

Model Description

Model: DistilBERT for Sequence Classification
Task: Binary Classification
Domain: Hospital/Healthcare Chatbot
Classes:
- True: User wants to switch branches
- False: General query/information seeking

Usage

from transformers import DistilBertTokenizer, DistilBertForSequenceClassification
import torch

# Load model and tokenizer
tokenizer = DistilBertTokenizer.from_pretrained("hitty28/branch-switch-classifier")
model = DistilBertForSequenceClassification.from_pretrained("hitty28/branch-switch-classifier")

# Predict
def predict(text):
    inputs = tokenizer(text, truncation=True, padding='max_length', max_length=128, return_tensors='pt')
    with torch.no_grad():
        outputs = model(**inputs)
        predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
        predicted_class = torch.argmax(predictions, dim=-1).item()
    return bool(predicted_class)

# Example
result = predict("I want to switch to Delhi branch")
print(result)  # True

Training Data

The model was trained on a comprehensive dataset including:

Direct branch switch requests
Location-specific switches
Facility-based switches
Information queries about branches
Medical service inquiries
Edge cases and ambiguous statements

Performance

The model achieves high accuracy in distinguishing between branch switching intents and general information queries.