Update app.py

app.py

@@ -536,67 +536,67 @@ def remove_incomplete_sentence(text):
     return text
 
 
-        language_code = 1
-        query_text = "recipes and meals for vegan diabetes headache fatigue"
-        print(f"Generated query text: {query_text}")
-
-        # Generate the query embedding
-        query_embedding = embed_query_text(query_text)
-        if query_embedding is None:
-            raise ValueError("Failed to generate query embedding.")
-
-        # Load embeddings and retrieve initial results
-        embeddings_data = load_recipes_embeddings()
-        folder_path = 'downloaded_articles/downloaded_articles'
-        initial_results = query_recipes_embeddings(query_embedding, embeddings_data, n_results=10)
-        if not initial_results:
-            raise ValueError("No relevant recipes found.")
-        print(initial_results)
-        # Extract document IDs
-        document_ids = [doc_id for doc_id, _ in initial_results]
-        print(document_ids)
-        # Retrieve document texts
-        document_texts = retrieve_rec_texts(document_ids, folder_path)
-        if not document_texts:
-            raise ValueError("Failed to retrieve document texts.")
-        print(document_texts)
-        # Load recipe metadata from DataFrame
-        folder_path='downloaded_articles/downloaded_articles'
-        file_path = 'recipes_metadata.xlsx'
-        metadata_path = 'recipes_metadata.xlsx'
-        metadata_df = pd.read_excel(file_path)
-        relevant_portions = extract_relevant_portions(document_texts, query_text, max_portions=3, portion_size=1, min_query_words=1)
-        print(relevant_portions)
-        flattened_relevant_portions = []
-        for doc_id, portions in relevant_portions.items():
-            flattened_relevant_portions.extend(portions)
-        unique_selected_parts = remove_duplicates(flattened_relevant_portions)
-        print(unique_selected_parts)
-        combined_parts = " ".join(unique_selected_parts)
-        print(combined_parts)
-        context = [query_text] + unique_selected_parts
-        print(context)
-        entities = extract_entities(query_text)
-        print(entities)
-        passage = enhance_passage_with_entities(combined_parts, entities)
-        print(passage)
-        prompt = create_prompt(query_text, passage)
-        print(prompt)
-        answer = generate_answer(prompt)
-        print(answer)
-        answer_part = answer.split("Answer:")[-1].strip()
-        print(answer_part)
-        cleaned_answer = remove_answer_prefix(answer_part)
-        print(cleaned_answer)
-        final_answer = remove_incomplete_sentence(cleaned_answer)
-        print(final_answer )
-        if language_code == 0:
-            final_answer = translate_en_to_ar(final_answer)
-        if final_answer:
-            print("Answer:")
-            print(final_answer)
-        else:
-            print("Sorry, I can't help with that.")
+language_code = 1
+query_text = "recipes and meals for vegan diabetes headache fatigue"
+print(f"Generated query text: {query_text}")
+
+# Generate the query embedding
+query_embedding = embed_query_text(query_text)
+if query_embedding is None:
+    raise ValueError("Failed to generate query embedding.")
+
+# Load embeddings and retrieve initial results
+embeddings_data = load_recipes_embeddings()
+folder_path = 'downloaded_articles/downloaded_articles'
+initial_results = query_recipes_embeddings(query_embedding, embeddings_data, n_results=10)
+if not initial_results:
+    raise ValueError("No relevant recipes found.")
+print(initial_results)
+# Extract document IDs
+document_ids = [doc_id for doc_id, _ in initial_results]
+print(document_ids)
+# Retrieve document texts
+document_texts = retrieve_rec_texts(document_ids, folder_path)
+if not document_texts:
+    raise ValueError("Failed to retrieve document texts.")
+print(document_texts)
+# Load recipe metadata from DataFrame
+folder_path='downloaded_articles/downloaded_articles'
+file_path = 'recipes_metadata.xlsx'
+metadata_path = 'recipes_metadata.xlsx'
+metadata_df = pd.read_excel(file_path)
+relevant_portions = extract_relevant_portions(document_texts, query_text, max_portions=3, portion_size=1, min_query_words=1)
+print(relevant_portions)
+flattened_relevant_portions = []
+for doc_id, portions in relevant_portions.items():
+    flattened_relevant_portions.extend(portions)
+unique_selected_parts = remove_duplicates(flattened_relevant_portions)
+print(unique_selected_parts)
+combined_parts = " ".join(unique_selected_parts)
+print(combined_parts)
+context = [query_text] + unique_selected_parts
+print(context)
+entities = extract_entities(query_text)
+print(entities)
+passage = enhance_passage_with_entities(combined_parts, entities)
+print(passage)
+prompt = create_prompt(query_text, passage)
+print(prompt)
+answer = generate_answer(prompt)
+print(answer)
+answer_part = answer.split("Answer:")[-1].strip()
+print(answer_part)
+cleaned_answer = remove_answer_prefix(answer_part)
+print(cleaned_answer)
+final_answer = remove_incomplete_sentence(cleaned_answer)
+print(final_answer )
+if language_code == 0:
+    final_answer = translate_en_to_ar(final_answer)
+if final_answer:
+    print("Answer:")
+    print(final_answer)
+else:
+    print("Sorry, I can't help with that.")