Templates (#12294)

Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
Co-authored-by: Lance Martin <lance@langchain.dev>
Co-authored-by: Jacob Lee <jacoblee93@gmail.com>

templates/rag-conversation/rag_conversation/chain.py

@@ -0,0 +1,92 @@
+from typing import Tuple, List
+from pydantic import BaseModel
+from operator import itemgetter
+from langchain.vectorstores import Pinecone
+from langchain.chat_models import ChatOpenAI
+from langchain.prompts import ChatPromptTemplate, MessagesPlaceholder
+from langchain.embeddings import OpenAIEmbeddings
+from langchain.schema import format_document, AIMessage, HumanMessage
+from langchain.prompts.prompt import PromptTemplate
+from langchain.schema.output_parser import StrOutputParser
+from langchain.schema.runnable import RunnablePassthrough, RunnableBranch, RunnableLambda, RunnableMap
+
+### Ingest code - you may need to run this the first time
+# Load
+# from langchain.document_loaders import WebBaseLoader
+# loader = WebBaseLoader("https://lilianweng.github.io/posts/2023-06-23-agent/")
+# data = loader.load()
+
+# # Split
+# from langchain.text_splitter import RecursiveCharacterTextSplitter
+# text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0)
+# all_splits = text_splitter.split_documents(data)
+#
+# # Add to vectorDB
+# vectorstore = Pinecone.from_documents(
+#     documents=all_splits, embedding=OpenAIEmbeddings(), index_name='langchain-test'
+# )
+# retriever = vectorstore.as_retriever()
+
+vectorstore = Pinecone.from_existing_index("langchain-test", OpenAIEmbeddings())
+retriever = vectorstore.as_retriever()
+
+# Condense a chat history and follow-up question into a standalone question
+_template = """Given the following conversation and a follow up question, rephrase the follow up question to be a standalone question, in its original language.
+Chat History:
+{chat_history}
+Follow Up Input: {question}
+Standalone question:"""
+CONDENSE_QUESTION_PROMPT = PromptTemplate.from_template(_template)
+
+# RAG answer synthesis prompt
+template = """Answer the question based only on the following context:
+<context>
+{context}
+</context>"""
+ANSWER_PROMPT = ChatPromptTemplate.from_messages([
+    ("system",template),
+    MessagesPlaceholder(variable_name="chat_history"),
+    ("user", "{question}")
+])
+
+# Conversational Retrieval Chain
+DEFAULT_DOCUMENT_PROMPT = PromptTemplate.from_template(template="{page_content}")
+def _combine_documents(docs, document_prompt = DEFAULT_DOCUMENT_PROMPT, document_separator="\n\n"):
+    doc_strings = [format_document(doc, document_prompt) for doc in docs]
+    return document_separator.join(doc_strings)
+
+def _format_chat_history(chat_history: List[Tuple[str, str]]) -> List:
+    buffer = []
+    for human, ai in chat_history:
+        buffer.append(HumanMessage(content=human))
+        buffer.append(AIMessage(content=ai))
+    return buffer
+
+# User input 
+class ChatHistory(BaseModel):
+    chat_history: List[Tuple[str, str]] 
+    question: str
+
+
+_search_query = RunnableBranch(
+        # If input includes chat_history, we condense it with the follow-up question
+        (
+            RunnableLambda(lambda x: bool(x.get("chat_history"))).with_config(
+                run_name="HasChatHistoryCheck"
+            ), # Condense follow-up question and chat into a standalone_question
+            RunnablePassthrough.assign(
+                chat_history=lambda x: _format_chat_history(x['chat_history'])
+            ) | CONDENSE_QUESTION_PROMPT | ChatOpenAI(temperature=0) | StrOutputParser(),
+        ),
+        # Else, we have no chat history, so just pass through the question
+        RunnableLambda(itemgetter("question"))
+
+    )
+
+_inputs = RunnableMap({
+    "question": lambda x: x["question"],
+    "chat_history": lambda x: _format_chat_history(x['chat_history']),
+    "context": _search_query | retriever | _combine_documents
+}).with_types(input_type=ChatHistory)
+
+chain = _inputs | ANSWER_PROMPT | ChatOpenAI() | StrOutputParser()