Update app.py

app.py

@@ -46,21 +46,11 @@ def get_document_store():
                                        index=doc_file_name)
     return document_store
 
-# @st.cache(allow_output_mutation=True)
-# def get_retriever(document_store):
-#     # Initialize the Pinecone retriever configuration here
-#     retriever = EmbeddingRetriever(
-#       document_store=document_store,
-#       embedding_model="sentence-transformers/multi-qa-mpnet-base-dot-v1",
-#       model_format="sentence_transformers",
-#       progress_bar=False,
-#     )
-#     return retriever
-
-# Get (or initialize and get) the document store and retriever
+
+# Get (or initialize and get) the document store
 document_store = get_document_store()
-# retriever = get_retriever(document_store)
 
+# Instantiate retriever
 retriever = EmbeddingRetriever(
   document_store=document_store,
   embedding_model="sentence-transformers/multi-qa-mpnet-base-dot-v1",
@@ -68,17 +58,6 @@ retriever = EmbeddingRetriever(
   progress_bar=False,
 )
 
-# template = PromptTemplate(
-#     prompt="""
-# Answer the given question using the following documents. \
-# Formulate your answer in the style of an academic report. \
-# Provide example quotes and citations using extracted text from the documents. \
-# Use facts and numbers from the documents in your answer. \
-# Reference information used from documents at the end of each applicable sentence (ex: [source: document_name]), where 'document_name' is the text provided at the start of each document (demarcated by '- &&&' and '&&&:')'. \
-# If no relevant information to answer the question is present in the documents, just say you don't have enough information to answer. \
-# Context: {' - '.join(['&&& '+d.meta['document']+' ref. '+str(d.meta['ref_id'])+' &&&: '+d.content for d in documents])}; Question: {query}; Answer:""",
-# )
-
 prompt_template="Answer the given question using the following documents. \
 Formulate your answer in the style of an academic report. \
 Provide example quotes and citations using extracted text from the documents. \
@@ -103,40 +82,6 @@ examples = [
     "In addition to gender, children, and youth, is there any mention of other groups facing disproportional impacts from climate change due to their geographic location, socio-economic status, age, gender, health, and occupation?"
 ]
 
-# def get_docs(input_query, country = None):
-#     '''
-#     Construct a hacky query to focus the retriever on the target country (see notes below)
-#     We take the top 150 k because we want to make sure we have 10 pertaining to the selected country
-#     '''
-#     if country == 'All Countries':
-#         query = input_query
-#     else:
-#         query = "For the country of "+country+", "+input_query
-#     # Retrieve top k documents
-#     docs = retriever.retrieve(query=query,top_k = 150)
-#     # Break out the key fields and convert to pandas for filtering
-#     docs = [{**x.meta,"score":x.score,"content":x.content} for x in docs]
-#     df_docs = pd.DataFrame(docs)
-#     if country != 'All Countries':
-#       df_docs = df_docs.query('country in @country')
-#     # Take the top 10
-#     df_docs = df_docs.head(10)
-#     # Get ourselves an index setup from which to base the source reference number from (in the prompt and matching afterwards)
-#     df_docs = df_docs.reset_index()
-#     df_docs['ref_id'] = df_docs.index + 1 # start the index at 1
-#     # Convert back to Document format
-#     ls_dict = []
-#     # Iterate over df and add relevant fields to the dict object
-#     for index, row in df_docs.iterrows():
-#         # Create a Document object for each row
-#         doc = Document(
-#             row['content'],
-#             meta={'country': row['country'],'document': row['document'], 'page': row['page'], 'file_name': row['file_name'], 'ref_id': row['ref_id'], 'score': row['score']}
-#         )
-
-#         # Append the Document object to the documents list
-#         ls_dict.append(doc)
-#     return(ls_dict)
 
 def get_docs(input_query, country = [], vulnerability_cat = []):
   if not country:
@@ -189,9 +134,9 @@ def get_refs(docs, res):
       ref_id = doc['meta']['ref_id']
       if ref_id in ref_ids:
         if doc['meta']['document'] == "Supplementary":
-            result_str += "**Ref. " + str(ref_id) + " [" + doc['meta']['country'] + " " + doc['meta']['document'] + ':' + doc['meta']['file_name'] + ' p' + str(doc['meta']['page']) + "]:** " + "*'" + doc['content'] + "'*<br> <br>" # Add <br> for a line break
+            result_str += "**Ref. " + str(ref_id) + " [" + doc['meta']['country'] + " " + doc['meta']['document'] + ':' + doc['meta']['file_name'] + ' p' + str(doc['meta']['page']) + '; vulnerabilities: ' + doc['meta']['vulnerability_cat'] + "]:** " + "*'" + doc['content'] + "'*<br> <br>" # Add <br> for a line break
         else:
-            result_str += "**Ref. " + str(ref_id) + " [" + doc['meta']['country'] + " " + doc['meta']['document'] + ' p' + str(doc['meta']['page']) + "]:** " + "*'" + doc['content'] + "'*<br> <br>" # Add <br> for a line break
+            result_str += "**Ref. " + str(ref_id) + " [" + doc['meta']['country'] + " " + doc['meta']['document'] + ' p' + str(doc['meta']['page']) + '; vulnerabilities: ' + doc['meta']['vulnerability_cat'] + "]:** " + "*'" + doc['content'] + "'*<br> <br>" # Add <br> for a line break
   
   return result_str
 
@@ -227,19 +172,15 @@ def run_query(input_text, country, model_sel):
 
 with st.sidebar:
     # Dropdown selectbox
-    country = st.sidebar.multiselect('Select a country:', country_options)
+    country = st.sidebar.multiselect('Filter by country:', country_options)
+    vulnerabilities_cat = st.sidebar.multiselect('Filter by vulnerabilities category:', vulnerability_options)
     st.markdown(
         """
-        * *For a comparative analysis of multiple countries, select **'All Countries'***
-        * *Then be sure to mention the country names of interest in your query*
+        * *These selections will filter the data matched against your query*
+        * *For a comparative analysis of multiple countries or vulnerability categories, select the items you require or select **'All Categories'***
+        * *Be careful in using the vulnerabilities category filter, as many of the categories are not well represented in the documents. Therefore, this will severly limit the data available for analysis*
         """
     )
-    vulnerabilities_cat = st.sidebar.multiselect('Select a vulnerabilities category:', vulnerability_options)
-    # choice = st.sidebar.radio(label = 'Select the Document',
-    #                         help = 'You can upload the document \
-    #                         or else you can try a example document', 
-    #                         options = ('Upload Document', 'Try Example'), 
-    #                         horizontal = True)
 
 with st.container():
         st.markdown("<h2 style='text-align: center;'> Climate Policy Documents: Vulnerabilities Analysis Q&A </h2>", unsafe_allow_html=True)
@@ -252,9 +193,16 @@ with st.expander("ℹ️ - About this app", expanded=False):
         
         **DISCLAIMER:** *This prototype tool based on LLMs (Language Models) is provided "as is" for experimental and exploratory purposes only, and should not be used for critical or production applications. Users are advised that the tool may contain errors, bugs, or limitations and should be used with caution and awareness of potential risks, and the developers make no warranties or guarantees regarding its performance, reliability, or suitability for any specific purpose.*
         """)
-
     # Display the text passages as radio buttons
     selected_example = st.radio("Example questions", examples)
+    st.write(
+    """
+    You can request comparative analyses between countries by filtering by country, and using more advanced prompts. For example:
+    
+    *Provide a comparative analysis between Angola and Kenya with regard to specific initiatives presented in the context to address the needs of groups such women and children to the effects climate change.*
+
+    Make sure your filters match the countries you have specified for the analysis!
+    """)