"""
Usage examples for LAPEFT Financial Sentiment Analysis
"""

from transformers import pipeline, BertTokenizer, BertForSequenceClassification
from peft import PeftModel
import torch

def simple_usage():
    """Simple pipeline usage - recommended for most users"""
    classifier = pipeline(
        "text-classification", 
        model="Hananguyen12/LAPEFT-Financial-Sentiment-Analysis"
    )
    
    # Example texts
    examples = [
        "The company exceeded earnings expectations with strong revenue growth.",
        "Market volatility continues with mixed signals from investors.",
        "Bankruptcy filing has caused significant concern among stakeholders."
    ]
    
    for text in examples:
        result = classifier(text)[0]
        print(f"Text: {text}")
        print(f"Sentiment: {result['label']} (Confidence: {result['score']:.3f})")
        print("-" * 50)

def advanced_usage():
    """Advanced usage with direct model access"""
    # Load model components
    base_model = BertForSequenceClassification.from_pretrained("bert-base-uncased", num_labels=3)
    model = PeftModel.from_pretrained(base_model, "Hananguyen12/LAPEFT-Financial-Sentiment-Analysis")
    tokenizer = BertTokenizer.from_pretrained("Hananguyen12/LAPEFT-Financial-Sentiment-Analysis")
    
    def predict_detailed(text):
        inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
        
        with torch.no_grad():
            outputs = model(**inputs)
            probs = torch.nn.functional.softmax(outputs.logits, dim=-1)
        
        return {
            "negative": probs[0][0].item(),
            "neutral": probs[0][1].item(),
            "positive": probs[0][2].item()
        }
    
    # Example
    text = "The quarterly report shows promising growth indicators."
    scores = predict_detailed(text)
    print(f"Text: {text}")
    print(f"Detailed scores: {scores}")

if __name__ == "__main__":
    print("=== Simple Usage ===")
    simple_usage()
    
    print("\n=== Advanced Usage ===")
    advanced_usage()