Update app.py

app.py

@@ -1,30 +1,58 @@
 import streamlit as st
 import requests
 import datetime
+import os
 
-API_URL = "https://api-inference.huggingface.co/models/mistralai/Mistral-7B-Instruct-v0.1"
-headers = {"Authorization": "Bearer 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"}
+from langchain.llms import HuggingFaceHub
 
-def query(payload):
-	response = requests.post(API_URL, headers=headers, json=payload, verify=False)
-	return response.json()
+from langchain.chains import RetrievalQA
 
-st.header("#CodeWars localGPT", divider='rainbow')
+# Embed and store
+from langchain.vectorstores import Chroma
+from langchain.embeddings import HuggingFaceHubEmbeddings
+from langchain.memory import ConversationBufferMemory
+from langchain.prompts import PromptTemplate
+
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.document_loaders import RecursiveUrlLoader
+from langchain.chains import RetrievalQA
+from langchain.document_loaders import PyPDFLoader
+
+os.environ["HUGGINGFACEHUB_API_TOKEN"] = "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"
+embedding = HuggingFaceHubEmbeddings()
+
+loader = PyPDFLoader("sample.pdf")
+pages = loader.load_and_split()
+
+vectorstore = Chroma.from_documents(documents=docs, embedding=embedding)
+
+
+memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True) # to remember chat history include this
 
-st.markdown("*#CodeWars* is **really** ***cool***... :sunglasses: our focus is on quality reasoning.")
+
+prompt_template = """Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.
+{context}
+Question: {question}"""
+
+PROMPT = PromptTemplate(
+   template=prompt_template, input_variables=["context", "question"]
+)
+chain_type_kwargs = {"prompt": PROMPT}
+
+llm = HuggingFaceHub(repo_id="mistralai/Mistral-7B-v0.1", model_kwargs={"temperature":0.1, "max_new_tokens":250})
+
+qachain=RetrievalQA.from_chain_type(llm, retriever=vectorstore.as_retriever(), memory=memory, chain_type_kwargs=chain_type_kwargs)
+st.header("#CodeWars localGPT", divider='rainbow')
 
 option = st.selectbox('What is your role?', ('Support', 'Sales'))
 
 st.write('You selected', option)
 
 prompt = st.chat_input("Say something to our #CodeWars bot...")
-if prompt:
-    
+context = [] # the context stores a conversation history, you can use this to make the model more context aware
+if(prompt):
     with st.chat_message(option):
          st.write(f"{datetime.datetime.now()} :red[{option}:] ", prompt)
+    context = qachain({"query": prompt})
     
-    response = query({
-        "inputs": prompt,
-    })
-
-    st.write(f"{datetime.datetime.now()}", response)
+    st.write(f"{datetime.datetime.now()}", context['result'])