Update app.py

app.py

@@ -138,6 +138,18 @@ def get_pub(x):
     except:
             return ' '
 
+def get_field(x):
+
+    try:
+        field = x['primary_topic']['subfield']['display_name']
+        if field is not None:
+            return field
+        else:
+            return np.nan
+    except:
+        return np.nan
+
+
 #def query_records(search_term):
 
 
@@ -269,6 +281,16 @@ def predict(text_input, sample_size_slider, reduce_sample_checkbox,sample_reduct
     stacked_df = pd.concat([basedata_df,records_df], axis=0, ignore_index=True)
     stacked_df = stacked_df.fillna("Unlabelled")
     stacked_df = stacked_df.reset_index(drop=True)
+
+
+    
+
+
+    stacked_df['parsed_field'] = [get_field(row) for ix, row in stacked_df.iterrows()]
+
+
+
+
     print(stacked_df)
 
     extra_data = pd.DataFrame(stacked_df['doi'])
@@ -320,7 +342,9 @@ def predict(text_input, sample_size_slider, reduce_sample_checkbox,sample_reduct
     
     plot = datamapplot.create_interactive_plot(
         stacked_df[['x','y']].values,
-        np.array(stacked_df['cluster_1_labels']),np.array(stacked_df['cluster_2_labels']),np.array(stacked_df['cluster_3_labels']),
+        # np.array(stacked_df['cluster_1_labels']),np.array(stacked_df['cluster_2_labels']),np.array(stacked_df['cluster_3_labels']),
+        np.array(['Unlabelled' if pd.isna(x) else x for x in stacked_df['parsed_field']]),
+
         hover_text=[str(row['title']) for ix, row in stacked_df.iterrows()],
         marker_color_array=stacked_df['color'],