Update app.py

app.py

@@ -104,7 +104,64 @@ with tab1:
 
 
     with col2:
-        st.subheader("Parking place near bikelane:")
+        st.subheader("Parking place near building:")
+
+        # Function to load the building models
+        
+        @st.cache_data()
+        def get_bikelane_mag_model(project=project):
+            mr = project.get_model_registry()
+            bikelane_mag_model = mr.get_model("building_mag_hist_model", version = 2)
+            bikelane_mag_model_dir = bikelane_mag_model.download()
+            return joblib.load(bikelane_mag_model_dir + "/bikelane_mag_hist_model.pkl")
+
+        # Retrieving model
+        bikelane_mag_hist_model = get_bikelane_mag_model()
+
+        @st.cache_data()
+        def get_bikelane_rad_model(project=project):
+            mr = project.get_model_registry()
+            bikelane_rad_model = mr.get_model("bikelane_rad_hist_model", version = 2)
+            bikelane_rad_model_dir = bikelane_rad_model.download()
+            return joblib.load(bikelane_rad_model_dir + "/bikelane_rad_hist_model.pkl")
+
+        # Retrieving model
+        bikelane_rad_hist_model = get_bikelane_rad_model()
+    
+        # Loading the feature group with latest data for bikelane
+        new_bikelane_fg = fs.get_feature_group(name = 'new_bikelane_fg', version = 1)
+
+        # Function to loading the feature group with latest data for bikelane as a dataset
+        @st.cache_data()
+        def retrieve_bikelane(feature_group=new_bikelane_fg):
+            new_bikelane_fg = feature_group.select_all()
+            df_bikelane_new = new_bikelane_fg.read(read_options={"use_hive": True})           
+            return df_bikelane_new
+
+        # Retrieving bikelane data
+        bikelane_new = retrieve_bikelane()
+
+        # Making the predictions and getting the latest data for magnetic field data
+        bikelane_mag_prediction_data = bikelane_new[['time', 'x', 'y', 'z', 'temperature', 'et0_fao_evapotranspiration']]                       
+        bikelane_mag_prediction_data['et0_fao_evapotranspiration'] = bikelane_mag_prediction_data['et0_fao_evapotranspiration'].apply(fill_nan_with_zero)
+        bikelane_mag_most_recent_prediction = bikelane_mag_prediction_data[['x', 'y', 'z', 'temperature', 'et0_fao_evapotranspiration']]
+        bikelane_mag_most_recent_prediction = bikelane_mag_hist_model.predict(bikelane_mag_most_recent_prediction)
+        bikelane_mag_prediction_data['Status'] = bikelane_mag_most_recent_prediction
+        bikelane_mag_prediction_data['Status'].replace(['detection', 'no_detection'], ['Vehicle detected', 'No vehicle detected'], inplace=True)
+        bikelane_mag_prediction_data = bikelane_mag_prediction_data.rename(columns={'time': 'Time'})
+        bikelane_mag_prediction_data = bikelane_mag_prediction_data.set_index(['Time'])
+        st.dataframe(bikelane_mag_prediction_data[['Status']].tail(3))
+
+        # Making the predictions and getting the latest data for radar data
+        bikelane_rad_prediction_data = bikelane_new[['time', 'radar_0', 'radar_1', 'radar_2', 'radar_3', 'radar_4', 'radar_5', 'radar_6', 'radar_7', 'temperature', 'et0_fao_evapotranspiration']]                       
+        bikelane_rad_prediction_data['et0_fao_evapotranspiration'] = bikelane_rad_prediction_data['et0_fao_evapotranspiration'].apply(fill_nan_with_zero)
+        bikelane_rad_most_recent_prediction = bikelane_rad_prediction_data[['radar_0', 'radar_1', 'radar_2', 'radar_3', 'radar_4', 'radar_5', 'radar_6', 'radar_7', 'temperature', 'et0_fao_evapotranspiration']]
+        bikelane_rad_most_recent_prediction = bikelane_rad_hist_model.predict(bikelane_rad_most_recent_prediction)
+        bikelane_rad_prediction_data['Status'] = bikelane_rad_most_recent_prediction
+        bikelane_rad_prediction_data['Status'].replace(['detection', 'no_detection'], ['Vehicle detected', 'No vehicle detected'], inplace=True)
+        bikelane_rad_prediction_data = bikelane_rad_prediction_data.rename(columns={'time': 'Time'})
+        bikelane_rad_prediction_data = bikelane_rad_prediction_data.set_index(['Time'])
+        st.dataframe(bikelane_rad_prediction_data[['Status']].tail(3))
             
 
     # Update button