Update app.py

app.py

@@ -229,8 +229,28 @@ with st.container():
 
 k_threshold = 5
 
+# Function to perform bootstrap on the predicted target values
+def bootstrap_stats(predicted_target, num_samples=1000):
+    # Reshape the predicted_target array
+    predicted_target = np.array(predicted_target).reshape(-1, 1)
+
+    # Bootstrap resampling
+    bootstrapped_means = []
+    for _ in range(num_samples):
+        bootstrap_sample = np.random.choice(predicted_target.flatten(), len(predicted_target), replace=True)
+        bootstrapped_means.append(np.mean(bootstrap_sample))
+
+    # Calculate lower and higher bounds
+    lower_bound = np.percentile(bootstrapped_means, 2.5)
+    higher_bound = np.percentile(bootstrapped_means, 97.5)
+
+    # Calculate the mean value
+    mean_value = np.mean(bootstrapped_means)
+
+    return lower_bound, higher_bound, mean_value
+
 # Apply KNN and get predicted Predicted_target values
-predicted_target = knn_predict(filtered_data, 'Predicted_target', ['latitude', 'longitude', 'area_feature'])  # Update with your features
+predicted_target = knn_predict(filtered_data, 'Predicted_target', ['latitude', 'longitude', 'area_feature'])
 
 # Check if there are predictions to display
 if 'Predicted_target' in filtered_data.columns and not np.all(predicted_target == 0):
@@ -240,5 +260,13 @@ if 'Predicted_target' in filtered_data.columns and not np.all(predicted_target =
     # Display the predicted Predicted_target values
     st.write("Valores (R$/m²) previstos com algoritmo KNN:")
     st.write(filtered_data[['Localização', 'Atotal', 'Apriv', 'Vunit_total', 'Vunit_priv', 'Predicted_target']])
+
+    # Apply bootstrap on the predicted values
+    lower_bound, higher_bound, mean_value = bootstrap_stats(predicted_target)
+
+    # Display the results
+    st.write(f"Valor médio (R$/m²) para as características selecionadas: {mean_value}")
+    st.write(f"Os valores podem variar entre {lower_bound} e {higher_bound} dependendo das características dos imóveis.")
+    st.write(f"Higher Bound: ")
 else:
     st.warning(f"Dados insuficientes para inferência do valor. Mínimo necessário: {k_threshold}")