Create app.py

app.py

@@ -0,0 +1,44 @@
+import gradio as gr
+import joblib
+import numpy as np
+
+# Load the trained model
+model = joblib.load("decision_tree_regressor_model.pkl")
+
+# Define the prediction function
+def predict_rpm(
+    TM,	U_percent,	Thin,	Thick,	Neps,	IPI,	HI,	ARKM,	RKM_CV,	MRKM,	Elongation,	Elon_CV	
+):
+    features = np.array([
+        TM,	U_percent,	Thin,	Thick,	Neps,	IPI,	HI,	ARKM,	RKM_CV,	MRKM,	Elongation,	Elon_CV
+    ]).reshape(1, -1)
+    prediction = model.predict(features)
+    return round(prediction[0], 2)
+
+# Define sliders for each feature
+inputs = [
+    gr.Slider(minimum=3.8, maximum=4.2, step=0.1, label="TM"),
+    gr.Slider(minimum=9.42, maximum=10.16, step=0.01, label="U_percent"),
+    gr.Slider(minimum=0.0, maximum=4, step=1, label="Thin"),
+    gr.Slider(minimum=18.0, maximum=40, step=1, label="Thick"),
+    gr.Slider(minimum=56, maximum=101.0, step=1, label="Neps"),
+    gr.Slider(minimum=89, maximum=122.0, step=1, label="IPI"),
+    gr.Slider(minimum=3.51, maximum=4.45, step=0.1, label="HI"),
+    gr.Slider(minimum=18.7, maximum=20.24, step=0.1, label="ARKM"),
+    gr.Slider(minimum=6.89, maximum=8.6, step=0.1, label="RKM_CV"),
+    gr.Slider(minimum=13.55, maximum=15.95, step=0.1, label="MRKM"),
+    gr.Slider(minimum=3.75, maximum=5.52, step=0.1, label="Elongation"),
+    gr.Slider(minimum=8.04, maximum=10.54, step=0.1, label="Elon_CV"),
+]
+
+# Create the interface
+demo = gr.Interface(
+    fn=predict_rpm,
+    inputs=inputs,
+    outputs=gr.Number(label="Predicted RPM"),
+    title="RPM Prediction App",
+    description="Adjust the sliders for each of the 12 features to predict the RPM."
+)
+
+if __name__ == "__main__":
+    demo.launch()