import gradio as gr
import pickle
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.preprocessing import LabelEncoder
from xgboost import XGBClassifier

# Load the model, label encoder, and vectorizer
with open('xgb_model.pkl', 'rb') as model_file:
    model = pickle.load(model_file)

with open('label_encoder.pkl', 'rb') as encoder_file:
    label_encoder = pickle.load(encoder_file)

with open('vectorizer.pkl', 'rb') as vectorizer_file:
    vectorizer = pickle.load(vectorizer_file)

# Define the prediction function
def predict(text):
    try:
        text_vector = vectorizer.transform([text])
        prediction = model.predict(text_vector)
        label = label_encoder.inverse_transform(prediction)[0]
        return {"prediction": label}
    except Exception as e:
        return {"error": str(e)}

# Create the Gradio interface
interface = gr.Interface(
    fn=predict,
    inputs=gr.Textbox(lines=2, placeholder="Enter a message..."),
    outputs="json",
    title="Spam Detector",
    description="Enter a message to determine if it is Phishing or Legitimate."
)

# Launch the Gradio app
interface.launch(share=True)