Create app.py

app.py

@@ -0,0 +1,59 @@
+import gradio as gr
+import joblib
+import pandas as pd
+from sklearn.preprocessing import StandardScaler, OneHotEncoder
+
+# Load the trained model
+model = joblib.load('logistic_regression_model.pkl')
+
+# Load the scaler and encoder used during training (if applicable)
+scaler = StandardScaler()  # Make sure to use the same scaler as during training
+encoder = OneHotEncoder(sparse=False, drop='first')  # Make sure to use the same encoder as during training
+
+# Define the input feature names
+input_features = [
+    "UDI",
+    "Air temperature [K]",
+    "Process temperature [K]",
+    "Rotational speed [rpm]",
+    "Torque [Nm]",
+    "Tool wear [min]"
+]
+
+def predict(UDI, air_temp, process_temp, rotational_speed, torque, tool_wear):
+    # Prepare input data
+    input_data = pd.DataFrame([[UDI, air_temp, process_temp, rotational_speed, torque, tool_wear]],
+                              columns=input_features)
+    
+    # Handle categorical features (if any) - Assuming 'UDI' might be categorical
+    non_numeric_columns = input_data.select_dtypes(exclude=['number']).columns
+    if len(non_numeric_columns) > 0:
+        input_data_encoded = encoder.transform(input_data[non_numeric_columns])
+        input_data_encoded = pd.DataFrame(input_data_encoded, index=input_data.index)
+        input_data = pd.concat([input_data.drop(non_numeric_columns, axis=1), input_data_encoded], axis=1)
+    
+    # Scale the features
+    input_data = scaler.transform(input_data)
+    
+    # Predict using the loaded model
+    prediction = model.predict(input_data)
+    return "Failure" if prediction[0] == 1 else "No Failure"
+
+# Create the Gradio interface
+interface = gr.Interface(
+    fn=predict,
+    inputs=[
+        gr.inputs.Number(label="UDI"),
+        gr.inputs.Number(label="Air temperature [K]"),
+        gr.inputs.Number(label="Process temperature [K]"),
+        gr.inputs.Number(label="Rotational speed [rpm]"),
+        gr.inputs.Number(label="Torque [Nm]"),
+        gr.inputs.Number(label="Tool wear [min]")
+    ],
+    outputs=gr.outputs.Textbox(label="Prediction"),
+    title="Predictive Maintenance Model",
+    description="Enter the values for the features to get the model's prediction."
+)
+
+# Launch the app
+interface.launch()