Update app.py

app.py

@@ -1,28 +1,43 @@
 import gradio as gr
+import pandas as pd
 import numpy as np
 import joblib
 import tensorflow as tf
 
-# Load local files 
+# Load model & scaler
 model = tf.keras.models.load_model("mlp_model.keras")
 scaler = joblib.load("scaler.pkl")
 
 N_FEATURES = model.input_shape[1]
 
-def predict(*features):
-    x = np.array(features).reshape(1, -1)
-    x_scaled = scaler.transform(x)
-    prob = model.predict(x_scaled)[0][0]
-    label = int(prob > 0.5)
-    return {
-        "probability": float(prob),
-        "prediction": label
-    }
-
-inputs = [gr.Number(label=f"Feature {i+1}") for i in range(N_FEATURES)]
-
-gr.Interface(
-    fn=predict,
-    inputs=inputs,
-    outputs="json"
-).launch()
+def predict_csv(file):
+    df = pd.read_csv(file)
+
+    # Check number of features
+    if df.shape[1] != N_FEATURES:
+        return f"Expected {N_FEATURES} features, but got {df.shape[1]} columns."
+
+    X = df.values.astype(float)
+    X_scaled = scaler.transform(X)
+    probs = model.predict(X_scaled).reshape(-1)
+    preds = (probs > 0.5).astype(int)
+    # Build result dataframe
+    result = df.copy()
+    result["probability_malware"] = probs
+    result["prediction"] = preds
+    result["prediction_label"] = result["prediction"].map(
+        {1: "malware", 0: "benign"}
+    )
+
+    return result
+
+
+demo = gr.Interface(
+    fn=predict_csv,
+    inputs=gr.File(label="Upload CSV file"),
+    outputs=gr.Dataframe(label="Prediction Result"),
+    title="Malware Detection MLP Model",
+    description="Upload a CSV file with features to predict malware or benign."
+)
+
+demo.launch()