Upload 2 files

app.py

@@ -7,21 +7,18 @@ from aimakerspace.openai_utils.prompts import (
     SystemRolePrompt,
     AssistantRolePrompt,
 )
-# from aimakerspace.openai_utils.embedding import EmbeddingModel
-# from aimakerspace.vectordatabase import VectorDatabase
-from langchain_openai import ChatOpenAI
-# from aimakerspace.openai_utils.chatmodel import ChatOpenAI
+from aimakerspace.openai_utils.embedding import EmbeddingModel
+from aimakerspace.vectordatabase import VectorDatabase
+from aimakerspace.openai_utils.chatmodel import ChatOpenAI
 import chainlit as cl
 from langchain_text_splitters import RecursiveCharacterTextSplitter
-# from langchain_experimental.text_splitter import SemanticChunker
-# from langchain_openai.embeddings import OpenAIEmbeddings
-from sentence_transformers import SentenceTransformer
+# from sentence_transformers import SentenceTransformer
 from langchain_huggingface import HuggingFaceEmbeddings
-from langchain_community.vectorstores import FAISS
-from langchain_openai.embeddings import OpenAIEmbeddings
-from langchain_core.documents import Document
+# from langchain_community.vectorstores import FAISS
+# from langchain_openai.embeddings import OpenAIEmbeddings
+# from langchain_core.documents import Document
 from dotenv import load_dotenv
-# from langchain.chains import RetrievalQA
+
 
 load_dotenv()
 
@@ -37,27 +34,27 @@ Question:
 """
 user_role_prompt = UserRolePrompt(user_prompt_template)
 
-# class RetrievalAugmentedQAPipeline:
-#     def __init__(self, llm: ChatOpenAI(), vector_db_retriever: VectorDatabase) -> None:
-#         self.llm = llm
-#         self.vector_db_retriever = vector_db_retriever
+class RetrievalAugmentedQAPipeline:
+    def __init__(self, llm: ChatOpenAI(), vector_db_retriever: VectorDatabase) -> None:
+        self.llm = llm
+        self.vector_db_retriever = vector_db_retriever
 
-#     async def arun_pipeline(self, user_query: str):
-#         context_list = self.vector_db_retriever.search_by_text(user_query, k=4)
+    async def arun_pipeline(self, user_query: str):
+        context_list = self.vector_db_retriever.search_by_text(user_query, k=4)
 
-#         context_prompt = ""
-#         for context in context_list:
-#             context_prompt += context[0] + "\n"
+        context_prompt = ""
+        for context in context_list:
+            context_prompt += context[0] + "\n"
 
-#         formatted_system_prompt = system_role_prompt.create_message()
+        formatted_system_prompt = system_role_prompt.create_message()
 
-#         formatted_user_prompt = user_role_prompt.create_message(question=user_query, context=context_prompt)
+        formatted_user_prompt = user_role_prompt.create_message(question=user_query, context=context_prompt)
 
-#         async def generate_response():
-#             async for chunk in self.llm.astream([formatted_system_prompt, formatted_user_prompt]):
-#                 yield chunk
+        async def generate_response():
+            async for chunk in self.llm.astream([formatted_system_prompt, formatted_user_prompt]):
+                yield chunk
 
-#         return {"response": generate_response(), "context": context_list}
+        return {"response": generate_response(), "context": context_list}
 
 text_splitter = RecursiveCharacterTextSplitter()
 
@@ -80,8 +77,8 @@ def process_text_file(file: AskFileResponse):
         documents = pdf_loader.load()
     else:
         raise ValueError("Provide a .txt or .pdf file")
-    #texts = [x.page_content for x in text_splitter.transform_documents(documents)]
-    # texts = [x.page_content for x in text_splitter.split_documents(documents)]
+    texts = [x.page_content for x in text_splitter.transform_documents(documents)]
+    
     return text_splitter.split_documents(documents)
 
 
@@ -100,7 +97,8 @@ async def on_chat_start():
             max_files=10
         ).send()
 
-    processed_documents = []
+    embedding_model = HuggingFaceEmbeddings(model_name='acpotts/finetuned_arctic')
+    vector_db = VectorDatabase(embedding_model=embedding_model)
     for file in files:
 
         msg = cl.Message(
@@ -110,50 +108,26 @@ async def on_chat_start():
 
         # load the file
         texts = process_text_file(file)
-        processed_documents.extend(texts)
+
         print(f"Processing {len(texts)} text chunks")
 
         # Create a dict vector store
-        # vector_db = VectorDatabase()
-        # vector_db = await vector_db.abuild_from_list(texts)
         
-    # chat_openai = ChatOpenAI()
+        vector_db = await vector_db.abuild_from_list(texts)
 
-    # Create a chain
-    # retrieval_augmented_qa_pipeline = RetrievalAugmentedQAPipeline(
-    #     vector_db_retriever=vector_db,
-    #     llm=chat_openai
-    # )
-    
-    # model = SentenceTransformer("acpotts/finetuned_arctic")
-    
-    finetune_embeddings = HuggingFaceEmbeddings(model_name='acpotts/finetuned_arctic')
+    chat_openai = ChatOpenAI()
 
-    finetune_vectorstore = FAISS.from_documents(processed_documents, finetune_embeddings)
-    finetune_retriever = finetune_vectorstore.as_retriever(search_kwargs={"k": 6})
-
-    from operator import itemgetter
-    from langchain_core.output_parsers import StrOutputParser
-    from langchain_core.runnables import RunnablePassthrough, RunnableParallel
-
-    rag_llm =  ChatOpenAI(
-        model="gpt-4o-mini",
-        temperature=0
-        )
-    
-
-    finetune_rag_chain = (
-        {"context": itemgetter("question") | finetune_retriever, "question": itemgetter("question")}
-        | RunnablePassthrough.assign(context=itemgetter("context"))
-        | {"response": system_template | rag_llm | StrOutputParser(), "context": itemgetter("context")}
+    #Create a chain
+    retrieval_augmented_qa_pipeline = RetrievalAugmentedQAPipeline(
+        vector_db_retriever=vector_db,
+        llm=chat_openai
     )
-
     
     # Let the user know that the system is ready
     msg.content = f"Processing `{file.name}` done. You can now ask questions!"
     await msg.update()
 
-    cl.user_session.set("chain", finetune_rag_chain)
+    cl.user_session.set("chain", retrieval_augmented_qa_pipeline)
 
 
 @cl.on_message
@@ -161,9 +135,7 @@ async def main(message):
     chain = cl.user_session.get("chain")
 
     msg = cl.Message(content="")
-    # finetune_rag_chain.invoke({"question": message.content})
-    # result = await chain.arun_pipeline(message.content)
-    result = await chain.arun_pipeline({'question': message.content})
+    result = await chain.arun_pipeline(message.content)
 
     async for stream_resp in result["response"]:
         await msg.stream_token(stream_resp)

requirements.txt

@@ -7,4 +7,3 @@ pypdf
 sentence_transformers
 langchain_text_splitters
 langchain-community
-faiss-cpu