Upload app.py

app.py

@@ -0,0 +1,50 @@
+import os
+import joblib
+import gradio as gr
+import pandas as pd
+
+# Load the pre-trained model
+price_predictor = joblib.load("model-v3.joblib")
+
+# Define the input components for the Gradio interface
+carat_input = gr.Number(label="Carat")
+shape_input = gr.Dropdown(['Round', 'Emerald', 'Marquise', 'Princess', 'Pear', 'Heart',
+                           'Oval', 'Cushion', 'Asscher', 'Radiant'], label='Shape')
+cut_input = gr.Dropdown(['Very Good', 'Ideal', 'Super Ideal', 'Good', 'Fair'], label='Cut')
+color_input = gr.Dropdown(['J', 'I', 'E', 'F', 'G', 'H', 'D'], label='Color')
+clarity_input = gr.Dropdown(['SI2', 'SI1', 'VS2', 'VVS1', 'VS1', 'VVS2', 'IF', 'FL'], label='Clarity')
+report_input = gr.Dropdown(['GIA', 'HRD', 'IGI', 'GCAL'], label='Report')
+type_input = gr.Dropdown(['natural', 'lab'], label='Type')
+
+# Define the output component for the Gradio interface
+model_output = gr.Label(label="Predicted Price (USD)")
+
+# Define the prediction function
+def predict_price(carat, shape, cut, color, clarity, report, type):
+    sample = {'carat': carat,  # Corrected key here
+              'shape': shape,
+              'cut': cut,
+              'color': color,
+              'clarity': clarity,
+              'report': report,
+              'type': type}
+    data_point = pd.DataFrame([sample])
+    prediction = price_predictor.predict(data_point).tolist()
+    return prediction[0]
+
+# Create the Gradio interface
+demo = gr.Interface(
+    fn=predict_price,
+    inputs=[carat_input, shape_input, cut_input, color_input, clarity_input, report_input, type_input],
+    outputs=model_output,
+    theme=gr.themes.Soft(),
+    title="Predictor of Diamond Valuations",
+    description="This application enables you to estimate the value of diamonds based on their characteristics",
+    # Uncomment the following lines if you have set up flagging
+    # allow_flagging="auto",
+    # flagging_callback=hf_writer,
+    concurrency_limit=8
+)
+
+# Launch the application
+demo.launch()