Create app.py

app.py

@@ -0,0 +1,42 @@
+import gradio as gr
+import pandas as pd
+import joblib
+
+# Load model
+model = joblib.load("predictive_model_smote.pkl")
+
+def predict_failure(type_value, air_temp, process_temp, rotational_speed, torque, tool_wear):
+    
+    # Create dataframe (same order as training)
+    input_data = pd.DataFrame([{
+        "Type": type_value,
+        "Air temperature [K]": air_temp,
+        "Process temperature [K]": process_temp,
+        "Rotational speed [rpm]": rotational_speed,
+        "Torque [Nm]": torque,
+        "Tool wear [min]": tool_wear
+    }])
+    
+    prediction = model.predict(input_data)[0]
+    
+    if prediction == 1:
+        return "⚠️ Machine Failure Likely"
+    else:
+        return "✅ Machine Working Normally"
+
+interface = gr.Interface(
+    fn=predict_failure,
+    inputs=[
+        gr.Dropdown(["L", "M", "H"], label="Type"),
+        gr.Number(label="Air Temperature (K)"),
+        gr.Number(label="Process Temperature (K)"),
+        gr.Number(label="Rotational Speed (rpm)"),
+        gr.Number(label="Torque (Nm)"),
+        gr.Number(label="Tool Wear (min)")
+    ],
+    outputs="text",
+    title="Predictive Maintenance Model",
+    description="Enter machine parameters to predict failure."
+)
+
+interface.launch()