Update app.py

app.py

@@ -1,68 +1,36 @@
 import gradio as gr
 import hopsworks
-import joblib
-import pandas as pd
 
-features = ['fixed_acidity',
-            'volatile_acidity',
-            'citric_acid',
-            'residual_sugar',
-            'chlorides',
-            'free_sulfur_dioxide',
-            'total_sulfur_dioxide',
-            'density',
-            'pH',
-            'sulphates',
-            'alcohol',
-            'is_white']
-labels = ["Low", "Medium", "High"]
+labels = ['Low', 'Medium', 'High']
 
 project = hopsworks.login()
 fs = project.get_feature_store()
 
-mr = project.get_model_registry()
-model = mr.get_model("wine_model", version=1)
-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_sulfur_dioxide, density, pH, sulphates, alcohol, white) -> str:
-    print("Calling function")
-    df = pd.DataFrame([[fixed_acidity, volatile_acidity, citric_acid, residual_sugar, chlorides, free_sulfur_dioxide,
-         total_sulfur_dioxide, density, pH, sulphates, alcohol, white]], columns=features)
-    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 f"{labels[res[0]]} quality"
-
-
-demo = gr.Interface(
-    fn=wine,
-    title="Wine Quality Predictive Analytics",
-    description="Experiment with wine characteristics to get the wine quality (low, medium, high)",
-    allow_flagging="never",
-    inputs=[
-        gr.components.Number(label='fixed acidity'),
-        gr.components.Number(label='volatile acidity'),
-        gr.components.Number(label='citric acid'),
-        gr.components.Number(label='residual sugar'),
-        gr.components.Number(label='chlorides'),
-        gr.components.Number(label='free sulfur dioxide'),
-        gr.components.Number(label='total sulfur dioxide'),
-        gr.components.Number(label='density'),
-        gr.components.Number(label='pH'),
-        gr.components.Number(label='sulphates'),
-        gr.components.Number(label='alcohol'),
-        gr.components.Checkbox(label='is white'),
-    ],
-    outputs=gr.Text())
-
-demo.launch(debug=True)
+dataset_api = project.get_dataset_api()
+dataset_api.download("Resources/images/wine_df_recent.png")
+dataset_api.download("Resources/images/wine_confusion_matrix.png")
+
+monitor_fg = fs.get_or_create_feature_group(name="wine_predictions", version=1, primary_key=["datetime"],
+                                            description="Wine quality Prediction/Outcome Monitoring")
+
+history_df = monitor_fg.read()
+last_prediction = history_df.tail()
+last_prediction = last_prediction[0] if len(last_prediction) > 0 else None
+
+with gr.Blocks() as demo:
+    with gr.Row():
+        with gr.Column():
+            gr.Label("Today's Predicted")
+            gr.Label(f"{labels[last_prediction['prediction']] + ' quality' if last_prediction is not None else 'No predictions yet'}")
+        with gr.Column():
+            gr.Label("Today's Actual quality")
+            gr.Label(f"{labels[last_prediction['label']] + ' quality' if last_prediction is not None else 'No predictions yet'}")
+    with gr.Row():
+        with gr.Column():
+            gr.Label("Recent Prediction History")
+            gr.Image("wine_df_recent.png", elem_id="recent-predictions")
+        with gr.Column():
+            gr.Label("Confusion Maxtrix with Historical Prediction Performance")
+            gr.Image("wine_confusion_matrix.png", elem_id="confusion-matrix")
+
+demo.launch()