Add application file

app.py

@@ -9,7 +9,7 @@ import numpy as np
 from io import BytesIO
 from great_expectations.dataset import PandasDataset
 from great_expectations.core import ExpectationSuite, ExpectationConfiguration
-from sklearn.metrics import confusion_matrix
+from sklearn.metrics import confusion_matrix, accuracy_score, f1_score, precision_score, recall_score
 import seaborn as sns
 import matplotlib.pyplot as plt
 api = '151p8WWCoctBzBeg.wRj1VwLA6wwjCS2aG7A51NsbhEbqVZ35wLl5g03b85EeetLKtpsO9bDOjy8DR2O3'
@@ -125,6 +125,13 @@ def wine_predict(fixed_acidity, volatile_acidity, citric_acid, residual_sugar,
     y_test = y_test[:num_samples_matrix]
     y_pred = model.predict(X_test)
     confusion_matrix_plot = plot_confusion_matrix(y_test, y_pred)
+    acc = accuracy_score(y_test, y_pred)
+    prec = precision_score(y_test, y_pred, average='weighted')
+    rec = recall_score(y_test, y_pred, average='weighted')
+    f1 = f1_score(y_test, y_pred, average='weighted')
+
+    pred_results = pd.DataFrame([['Random Forest', acc, prec, rec, f1]],
+                columns = ['Model', 'Accuracy', 'Precision', 'Recall', 'F1 Score'])
 
     if num_samples == -1:
         df = pd.DataFrame([[fixed_acidity, volatile_acidity, citric_acid, residual_sugar, 
@@ -147,27 +154,13 @@ def wine_predict(fixed_acidity, volatile_acidity, citric_acid, residual_sugar,
         synthetic_wines_df = generate_samples(num_samples)  # This will also insert the data into the feature store
         synthetic_wines_df = synthetic_wines_df.drop('quality', axis=1)
         y_pred_new = model.predict(synthetic_wines_df)
-        synthetic_wines_df = synthetic_wines_df[:100]
-        wine_fg_50 = fs.get_feature_group(name="wine_2", version=1)
-        query = wine_fg_50.select_all()
-        feature_view = fs.get_or_create_feature_view(name="wine_2",
-                                        version=1,
-                                        description="Read from Wine Quality dataset",
-                                        labels=["quality"],
-                                        query=query)
-
-        X_train, X_test, y_train, y_test = feature_view.train_test_split(test_size = 0.2)
-        num_samples_matrix = int(num_samples_matrix)
-        X_test = X_test[:num_samples_matrix]
-        y_test = y_test[:num_samples_matrix]
-        y_pred = model.predict(X_test)
-        confusion_matrix_plot = plot_confusion_matrix(y_test, y_pred)
+        synthetic_wines_df = synthetic_wines_df[:50]
         res = y_pred_new.mean()
 
     if isinstance(synthetic_wines_df, pd.DataFrame):
-        return res, synthetic_wines_df, confusion_matrix_plot
+        return res, synthetic_wines_df, confusion_matrix_plot,pred_results
     else:
-        return res, pd.DataFrame(synthetic_wines_df), confusion_matrix_plot
+        return res, pd.DataFrame(synthetic_wines_df), confusion_matrix_plot,pred_results
 
     
 def update_dashboard(fixed_acidity, volatile_acidity, citric_acid, residual_sugar, 
@@ -175,13 +168,13 @@ def update_dashboard(fixed_acidity, volatile_acidity, citric_acid, residual_suga
          ph, sulphates, alcohol, num_samples, num_samples_matrix):
     
 
-    predction ,generated_samples,confusion_matrix_plot = wine_predict(fixed_acidity, 
+    predction ,generated_samples,confusion_matrix_plot,pred_results= wine_predict(fixed_acidity, 
                                                                         volatile_acidity, citric_acid, residual_sugar, 
                                                                         chlorides, free_sulfur_dioxide, total_sulfur_dioxide, density, 
                                                                         ph, sulphates, alcohol, num_samples, num_samples_matrix)
     
 
-    return predction, confusion_matrix_plot, generated_samples
+    return predction, confusion_matrix_plot, pred_results,generated_samples
 demo = gr.Interface(
     fn=update_dashboard,
     title="Wine Quality Predictive Analytics",
@@ -205,6 +198,7 @@ demo = gr.Interface(
     outputs=[
             gr.outputs.Textbox(label="Predicted Quality"),
             gr.outputs.Image(label="Confusion Matrix",type="pil"),
+            gr.outputs.Dataframe(label="Most Recent Prediction results",type='pandas'),
             gr.outputs.Dataframe(label="Most Recent Generated Wine Data",type='pandas')
             ],
 )