model..py

from transformers import pipeline

# Function to load the model
def load_model(model_name):
    try:
        # Load the model from Hugging Face or local storage (by name)
        model = pipeline("text-classification", model=model_name)
        return model
    except Exception as e:
        print(f"Error loading model: {e}")
        return None

# Function to run inference using the selected model
def run_inference(user_input, selected_model, prompt=None):
    model = load_model(selected_model)
    if model:
        # If a prompt is provided, prepend it to the input text
        if prompt:
            input_text = f"{prompt}\n{user_input}"
        else:
            input_text = user_input
        
        try:
            # Run inference and check model output
            result = model(input_text)
            
            # Assuming the output format is a list of dicts with 'label' field
            return result[0]['label'] if 'label' in result[0] else "Error: No label in output"
        except Exception as e:
            return f"Error during inference: {e}"
    else:
        return f"Error: Model '{selected_model}' failed to load."

# Example usage
selected_model = "Canstralian/CySec_Known_Exploit_Analyzer"
user_input = "Sample exploit description"
prompt = "Classify the following cybersecurity exploit:"

# Run inference
result = run_inference(user_input, selected_model, prompt)
print(f"Inference Result: {result}")