Upload app.py

app.py

@@ -0,0 +1,77 @@
+import modal
+from PIL import Image
+
+MODEL_VERSION = 1
+LOCAL = False
+
+if LOCAL == False:
+   hopsworks_image = modal.Image.debian_slim(python_version='3.9').pip_install(["gradio", "requests", "hopsworks", "joblib", "pandas", "scikit-learn==1.1.1"])
+   stub = modal.Stub("wine_prediction_user_interface")
+   @stub.function(image=hopsworks_image, secret=modal.Secret.from_name("HOPSWORKS_API_KEY"))
+   def f():
+       g()
+
+def g():
+    import gradio as gr
+    import requests
+    import hopsworks
+    import joblib
+    import pandas as pd
+
+    project = hopsworks.login()
+    fs = project.get_feature_store()
+
+    mr = project.get_model_registry()
+    model = mr.get_model("wine_model", version=MODEL_VERSION)
+    model_dir = model.download()
+    model = joblib.load(model_dir + "/wine_model.pkl")
+    print("Model downloaded")
+
+    def wine(fixed_acidity, volatile_acidity, citric_acid, residual_sugar, chlorides, free_sulfur_dioxide,
+            total_sulfar_dioxide, density, ph, sulphates, alcohol, color):
+        print("Calling function")
+    #     df = pd.DataFrame([[sepal_length],[sepal_width],[petal_length],[petal_width]], 
+        df = pd.DataFrame([[fixed_acidity, volatile_acidity, citric_acid, residual_sugar, chlorides, free_sulfur_dioxide,
+                            total_sulfar_dioxide, density, ph, sulphates, alcohol, color]], 
+                        columns=['fixed_acidity', 'volatile_acidity', 'citric_aicd', 'residual_sugar', 'chlorides', 'free_sulfur_dioxide',
+                            'total_sulfur_dioxide', 'density', 'ph', 'sulphates', 'alcohol', 'color'])
+        print("Predicting")
+        print(df)
+        # 'res' is a list of predictions returned as the label.
+        res = model.predict(df)
+        # We add '[0]' to the result of the transformed 'res', because 'res' is a list, and we only want 
+        # the first element.
+    #     print("Res: {0}").format(res)
+        print(res)
+        return res
+            
+    demo = gr.Interface(
+        fn=wine,
+        title="Wine Quality Predictive Analytics",
+        description="Experiment with several parameters to predict the quality of wine.",
+        allow_flagging="never",
+        inputs=[
+            gr.Number(value=1.0, label="fixed_acidity"),
+            gr.Number(value=1.0, label="volatile_acidity"),
+            gr.Number(value=1.0, label="citric_aicd"),
+            gr.Number(value=1.0, label="residual_sugar"),
+            gr.Number(value=1.0, label="chlorides"),
+            gr.Number(value=1.0, label="free_sulfur_dioxide"),
+            gr.Number(value=1.0, label="total_sulfur_dioxide"),
+            gr.Number(value=1.0, label="density"),
+            gr.Number(value=1.0, label="ph"),
+            gr.Number(value=1.0, label="sulphates"),
+            gr.Number(value=1.0, label="alcohol"),
+            gr.Number(value=1.0, label="color (0:red, 1:white)"),
+            ],
+        outputs=gr.Number(label="predicted quality"))
+
+    demo.launch(debug=True, share=True)
+
+if __name__ == "__main__":
+    if LOCAL == True :
+        g()
+    else:
+        modal.runner.deploy_stub(stub)
+        with stub.run():
+            f.remote()