added history

app.py

@@ -1,30 +1,16 @@
 import gradio as gr
 import pandas as pd
 from langchain.embeddings import HuggingFaceEmbeddings
-from langchain.vectorstores import Chroma, faiss
-from langchain_community.llms import HuggingFaceEndpoint, HuggingFaceHub
-from langchain.chains import LLMChain
-from langchain_community.document_loaders.csv_loader import CSVLoader
+from langchain.vectorstores import faiss
+from langchain_community.llms import HuggingFaceHub
+from langchain.chains import ConversationalRetrievalChain, LLMChain
 from langchain_community.document_loaders import PyPDFLoader
-from langchain.text_splitter import CharacterTextSplitter
-from langchain_community.document_loaders import TextLoader
-from langchain_community import vectorstores
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain import vectorstores
 from langchain.prompts import PromptTemplate
-from langchain.chains import RetrievalQA
 from langchain.memory import ConversationBufferMemory
-from langchain.chains import ConversationalRetrievalChain
-from langchain.text_splitter import CharacterTextSplitter, RecursiveCharacterTextSplitter
-from langchain.vectorstores import DocArrayInMemorySearch
-from langchain.document_loaders import TextLoader
-from langchain.chains import RetrievalQA,  ConversationalRetrievalChain
-from langchain.memory import ConversationBufferMemory
-from langchain.chat_models import ChatOpenAI
-from langchain.document_loaders import TextLoader
-from langchain.document_loaders import PyPDFLoader
-import panel as pn
-import param
-import re
 import os
+import re
 
 api_token = os.environ.get('HUGGINGFACEHUB_API_TOKEN')
 
@@ -67,19 +53,39 @@ def load_db(file, k):
         retriever=retriever,
         return_source_documents=True,
         return_generated_question=True,
+        memory=memory,
     )
 
     return qa
 
-def greet(question, pdf_file):
-    a = load_db(pdf_file, 3)
-    r = a.invoke({"question": question, "chat_history": []})
-    match = re.search(r'Helpful Answer:(.*)', r['answer'])
+def chat(input_text, pdf_file):
+    qa = load_db(pdf_file, 3)
+    
+    if not memory.memory:
+        # If no previous conversation, start with a greeting
+        response = qa.invoke({"question": "Hi, how can I help you today?", "chat_history": []})
+        memory.update(response["chat_history"])
+
+    response = qa.invoke({"question": input_text, "chat_history": memory.memory})
+    
+    # Extracting the helpful answer from the response
+    match = re.search(r'Helpful Answer:(.*)', response['answer'])
     if match:
         helpful_answer = match.group(1).strip()
-        return helpful_answer
     else:
-        return "No helpful answer found."
+        helpful_answer = "No helpful answer found."
+
+    # Update the chat history
+    memory.update([(input_text, helpful_answer)])
+
+    # Combine relevant information into a single string for output
+    output_text = f"Question: {input_text}\nAnswer: {helpful_answer}\nGenerated Question: {response['generated_question']}\nSource Documents: {response['source_documents']}"
 
-iface = gr.Interface(fn=greet, inputs=["text", "file"], outputs="text")
+    return output_text
+
+
+
+iface = gr.Interface(fn=chat, inputs=["text", "file"], outputs="text")
 iface.launch(share=True)
+
+