prediction with manual range selection

src/pages/3-prediction.py

@@ -137,8 +137,9 @@ with c1:
             x_block.columns = x_block.columns.astype(str)
             yname = system_data['data']['target'].name
             st.info("Loaded model to predict " + yname)
-            if ObjectHash(current=None, add=list(system_data['predictors_'])) == ObjectHash(current=None, add=list(x_block.columns)):
-                pred_data = x_block
+            shared_elements = set(system_data['predictors_']).intersection(x_block.columns)
+            if len(shared_elements)==len(system_data['predictors_']):
+                pred_data = x_block.loc[:,system_data['predictors_']]
             else:
                 st.error(
                     'The names of the features (columns) in the training set and the prediction set are not identical. Thus, prediction cannot be performed.')
@@ -237,13 +238,11 @@ if not preprocessed.empty:
         st.session_state['predict'] = True
 
     if not preprocessed.empty and st.session_state['predict']:
-        c4, c5 = st.columns([2, 1])
-        with c5:
-            model = system_data['model_']
-            if system_data['model_type'] in ['PLS', 'TPE-iPLS']:
-                nvar = system_data['model_'].n_features_in_
-            elif system_data['model_type'] == 'LW-PLS':
-                nvar = system_data['data']['raw-spectra'].shape[1]
+        model = system_data['model_']
+        if system_data['model_type'] in ['PLS', 'TPE-iPLS']:
+            nvar = system_data['model_'].n_features_in_
+        elif system_data['model_type'] == 'LW-PLS':
+            nvar = system_data['data']['raw-spectra'].shape[1]
 
         if system_data['selected-wls'] is None:
             preprocesseddf = preprocessed
@@ -317,11 +316,12 @@ if not preprocessed.empty:
             ################################### results display ###################################
 
 if not result.empty:
+    c4, c5 = st.columns([2, 1])
     hist = pred_hist(pred=result)
     with c4:
         st.write('Predicted values distribution')
         st.pyplot(hist)
-        st.write('Predicted values table')
+        st.write('Table of predicted values')
         resultT = result.reset_index()
         st.dataframe(resultT.T)
     with c5: