Update app.py

app.py

@@ -1,50 +1,72 @@
-from cgitb import text
-import streamlit as st
-from dotenv import load_dotenv
-from PyPDF2 import PdfReader
-from langchain_text_splitters import CharacterTextSplitter
-from langchain.embeddings import HuggingFaceInstructEmbeddings
-from langchain.vectorstores import FAISS 
-
-
-def get_text_chunks(raw_text):
-    text_splitter = CharacterTextSplitter(
-        separator="\n",
-        chunk_size=1000, chunk_overlap=200,
-        length_function= len
-    )
-    chunks = text_splitter.split(text)
-    return chunks
-    
-def get_pdf_text(pdf_docs):
-    text =""
-    for pdf in pdf_docs:
-        pdf_reader = PdfReader(pdf)
-        for page in pdf_reader.pages:
-            text += page.extract_text()
-    return text 
-
-def get_vectorestore(text_chunks):
-    embeddings = 
-    return 
-         
-    
-def main():
-    load_dotenv()
-    st.set_page_config(page_title="Chat with your PDFs", page_icon= ":books:")
-    st.header("Chat with your PDFs :books:")
-    st.text_input("Ask a question about your documents:")
-    with st.sidebar:
-        st.subheader("Your Documemts")
-        pdf_docs = st.file_uploader("Upload your PDFs here", accept_multiple_files= True)
-        if st.button("Enter"):
-            with st.spinner("Processing"):
-            # get pdf text
-                raw_text = get_pdf_text(pdf_docs)
-            # get text chunks
-                text_chunks = get_text_chunks(raw_text)
-            # create vector store
-                vectorstore = get_vectorstore(text_chunks)
-        
-if __name__ == "__main__":
+from cgitb import text
+import streamlit as st
+from dotenv import load_dotenv
+from PyPDF2 import PdfReader
+from langchain_text_splitters import CharacterTextSplitter
+from langchain.embeddings import HuggingFaceInstructEmbeddings
+from langchain.vectorstores import FAISS 
+from langchain.memory import ConversationalBufferMemory
+from langchain.chains import ConversationalRetrievalChain
+from langchain.llms import HuggingFaceHub
+
+
+def get_text_chunks(raw_text):
+    text_splitter = CharacterTextSplitter(
+        separator="\n",
+        chunk_size=1000, chunk_overlap=200,
+        length_function= len
+    )
+    chunks = text_splitter.split(text)
+    return chunks
+    
+def get_pdf_text(pdf_docs):
+    text =""
+    for pdf in pdf_docs:
+        pdf_reader = PdfReader(pdf)
+        for page in pdf_reader.pages:
+            text += page.extract_text()
+    return text 
+
+def get_vectorestore(text_chunks):
+    embeddings = HuggingFaceInstructEmbeddings(model_name= "hkunlp/instructor-xl")
+    vectorestore = FAISS.from_texts(texts=text_chunks, embedding= embeddings)
+    return vectorestore
+         
+ def get_conversation_chain(vectorstore):
+         llm = HuggingFaceHub(repo_id="google/flan-t5-xxl", model_kwargs={"temperature":0.5, "max_length":512})
+         memory = ConversationalBufferMemory(memory_key="chat_history", return_message= True)
+         conversation_chain = ConversationalRetrievalChain.from_llm(
+             llm=llm,
+             retriever=vectorstore.as_retriever(),
+             memory= memory
+         )
+     return get_conversation_chain
+     
+     
+     return 
+def main():
+    load_dotenv()
+    st.set_page_config(page_title="Chat with your PDFs", page_icon= ":books:")
+    
+    if "conversation" not in st.session_state:
+        st.session_state.conversation = None
+        
+    st.header("Chat with your PDFs :books:")
+    st.text_input("Ask a question about your documents:")
+    with st.sidebar:
+        st.subheader("Your Documemts")
+        pdf_docs = st.file_uploader("Upload your PDFs here", accept_multiple_files= True)
+        if st.button("Enter"):
+            with st.spinner("Processing"):
+            # get pdf text
+                raw_text = get_pdf_text(pdf_docs)
+            # get text chunks
+                text_chunks = get_text_chunks(raw_text)
+            # create vector store
+                vectorstore = get_vectorstore(text_chunks)
+                
+            # conversation_chain
+                st.session_state.conversation = get_conversation_chain(vectorstore)
+    st.session_state.conversation    
+if __name__ == "__main__":
     main()