experiment

functions/calc_matches.py

@@ -50,6 +50,10 @@ def calc_matches(filtered_df, project_df, similarity_matrix, top_x):
     filtered_df_indices = filtered_df.index.to_list()
     project_df_indices = project_df.index.to_list()
 
+    # Create mapping dictionaries
+    filtered_df_index_map = {index: i for i, index in enumerate(filtered_df_indices)}
+    project_df_index_map = {index: i for i, index in enumerate(project_df_indices)}
+
     st.write(filtered_df_indices)
     st.write(project_df_indices)
 
@@ -70,13 +74,16 @@ def calc_matches(filtered_df, project_df, similarity_matrix, top_x):
     # Get the corresponding similarity values
     top_values = match_matrix.data[linear_indices]
 
+    top_filtered_df_indices = [filtered_df_indices[i] for i in top_indices[0]]
+    top_project_df_indices = [project_df_indices[i] for i in top_indices[1]]
+
     # Create resulting dataframes with top matches and their similarity scores
     st.write(top_indices)
-    p1_df = filtered_df.loc[top_indices[0]].copy()
+    p1_df = filtered_df.loc[top_filtered_df_indices].copy()
     p1_df['similarity'] = top_values
 
     st.dataframe(p1_df)
-    p2_df = project_df.loc[top_indices[1]].copy()
+    p2_df = project_df.loc[top_project_df_indices].copy()
     p2_df['similarity'] = top_values
     st.dataframe(p2_df)
     print("finished calc matches")