Update functions.py

functions.py

@@ -30,7 +30,7 @@ from langchain.llms import OpenAI
 from langchain.callbacks.base import CallbackManager
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 from langchain.chains.chat_vector_db.prompts import CONDENSE_QUESTION_PROMPT
-from langchain import VectorDBQAWithSourcesChain
+from langchain import VectorDBQA
 
 
 from langchain.chat_models import ChatOpenAI
@@ -67,23 +67,29 @@ output_parser = RegexParser(
     output_keys=["answer", "score"],
 )
 
-system_template = """Given the following extracted parts of a long document and a question, create a final answer with references ("SOURCES"). 
-If you don't know the answer, just say that you don't know. Don't try to make up an answer.
+system_template="""Use only the following pieces of finance context to answer the users question thoroughly.
+Do not use any information not provided in the context. 
+If you don't know the answer, just say that you don't know, don't try to make up an answer.
 ALWAYS return a "SOURCES" part in your answer.
+The "SOURCES" part should be a reference to the source of the document from which you got your answer.
 
-In addition to giving an answer, also return a score of how fully it answered the user's question. This should be in the following format:
+Example of your response should be:
 
-Question: [question here]
-Helpful Answer: [answer here]
-Score: [score between 0 and 100]
+```
+The answer is foo
+SOURCES: xyz
+```
 
-Begin!
+If there is no sources found please return the below:
 
-Context:
----------
-{summaries}
----------
-"""
+```
+The answer is foo
+SOURCES: Please refer to references section
+```
+
+Begin!
+----------------
+{context}"""
 
 messages = [
     SystemMessagePromptTemplate.from_template(system_template),
@@ -91,33 +97,6 @@ messages = [
 ]
 prompt = ChatPromptTemplate.from_messages(messages)
 
-#Refine Chain Type Prompt Template
-refine_prompt_template = (
-    "The original question is as follows: {question}\n"
-    "We have provided an existing answer: {existing_answer}\n"
-    "We have the opportunity to refine the existing answer"
-    "(only if needed) with some more context below.\n"
-    "------------\n"
-    "{context_str}\n"
-    "------------\n"
-    "Given the new context, refine the original answer to better "
-    "answer the question. "
-    "If the context isn't useful, return the original answer."
-)
-refine_prompt = PromptTemplate(
-    input_variables=["question", "existing_answer", "context_str"],
-    template=refine_prompt_template,
-)
-
-initial_qa_template = (
-    "Context information is below. \n"
-    "---------------------\n"
-    "{context_str}"
-    "\n---------------------\n"
-    "Given the context information and not prior knowledge, "
-    "answer the question: {question}\n.\n"
-)
-
 ###################### Functions #######################################################################################
 
 @st.experimental_singleton(suppress_st_warning=True)
@@ -204,52 +183,27 @@ def gen_embeddings(embedding_model):
     return embeddings
     
 @st.experimental_memo(suppress_st_warning=True)
-def embed_text(query,title,embedding_model,_docsearch,chain_type):
+def embed_text(query,title,embedding_model,_docsearch):
     
     '''Embed text and generate semantic search scores'''
 
-    llm = OpenAI(temperature=0)
+    # llm = OpenAI(temperature=0)
     chat_llm = ChatOpenAI(streaming=True, callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]), verbose=True, temperature=0)
 
     title = title.split()[0].lower()
-                
-    docs = _docsearch.similarity_search_with_score(query, k=3)
-
-    if chain_type == 'Normal':
 
-        docs = [d[0] for d in docs]
+    chain_type_kwargs = {"prompt": prompt}
 
-        # PROMPT = PromptTemplate(template=template, 
-        #                         input_variables=["summaries", "question"],
-        #                         output_parser=output_parser)
-
-        chain_type_kwargs = {"prompt": prompt}
-        chain = VectorDBQAWithSourcesChain.from_chain_type(
-        streaming_llm, 
+    chain = VectorDBQA.from_chain_type(
+        llm=streaming_llm, 
         chain_type="stuff", 
-        vectorstore=_docsearch,
-        chain_type_kwargs=chain_type_kwargs
-    )
-        answer = chain({"question": query}, return_only_outputs=True)
-        # chain = load_qa_with_sources_chain(OpenAI(temperature=0), 
-        #                                    chain_type="stuff", 
-        #                                    prompt=PROMPT, 
-        #                                    )
-
-        
-        # answer = chain({"input_documents": docs, "question": query}, return_only_outputs=False)
-
-
-    elif chain_type == 'Refined':
-
-        docs = [d[0] for d in docs]
-        
-        initial_qa_prompt = PromptTemplate(
-    input_variables=["context_str", "question"], template=initial_qa_template
-)
-        chain = load_qa_chain(OpenAI(temperature=0), chain_type="refine", return_refine_steps=False,
-                     question_prompt=initial_qa_prompt, refine_prompt=refine_prompt)
-        answer = chain({"input_documents": docs, "question": query}, return_only_outputs=False)
+        vectorstore=_docsearch, 
+        chain_type_kwargs=chain_type_kwargs,
+        return_source_documents=True,
+        k=3
+        )
+    
+    answer = chain({"query": query})
 
     return answer