app.py

import joblib
import pandas as pd
import gradio as gr
import numpy as np

# ======================
# LOAD MODEL
# ======================
artifact = joblib.load("stacking_model.pkl")

base_models = artifact["base_models"]      # list of (name, model)
meta_model = artifact["meta_model"]
feature_names = artifact["features"]

# ======================
# PREDICTION FUNCTION
# ======================
def predict_malware(*inputs):
    # Input → DataFrame
    X = pd.DataFrame([inputs], columns=feature_names)

    # Level-1 predictions
    meta_inputs = []

    for name, model in base_models:
        prob = model.predict_proba(X)[:, 1]
        meta_inputs.append(prob)

    meta_X = np.column_stack(meta_inputs)

    # Meta prediction
    pred = meta_model.predict(meta_X)[0]
    prob = meta_model.predict_proba(meta_X)[0, 1]

    label = "Malware" if pred == 1 else "Benign"
    return label, float(prob)

# ======================
# UI
# ======================
inputs = [gr.Number(label=feat) for feat in feature_names]

outputs = [
    gr.Textbox(label="Prediction"),
    gr.Number(label="Malware Probability")
]

app = gr.Interface(
    fn=predict_malware,
    inputs=inputs,
    outputs=outputs,
    title="Stacking-based Malware Detection",
    description="ExtraTrees + RandomForest + LightGBM + LogisticRegression → XGBoost"
)

if __name__ == "__main__":
    app.launch()