Update app.py

app.py

@@ -1,54 +1,67 @@
+# generic libraries
 import gradio as gr
-from huggingface_hub import InferenceClient
+import os
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# for embeddings and indexing
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+
+# for data retrieval
+from langchain.chains import RetrievalQA
+
+# for huggingface llms
+from langchain_community.llms import HuggingFaceHub
+
+# define constants
+EMB_MODEL1 = 'BAAI/bge-base-en-v1.5'
+MISTRAL_MODEL1 = 'mistralai/Mixtral-8x7B-Instruct-v0.1'
+HF_MODEL1 = 'HuggingFaceH4/zephyr-7b-beta'
+# define paths
+vector_path = 'faiss_index'
+
+# Initialize your embedding model
+embedding_model = HuggingFaceEmbeddings(model_name=EMB_MODEL1)
+
+# Load FAISS from relative path
+if os.path.exists("faiss_index"):
+    vectordb = FAISS.load_local(vector_path, embedding_model)
+else:
+    raise FileNotFoundError("FAISS index not found in Space. Please upload it to faiss_index/")
 
 
 def respond(
     message,
-    history: list[tuple[str, str]],
-    system_message,
+    #history: list[tuple[str, str]],
+    #system_message,
     max_tokens,
     temperature,
     top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
+    vectordb,
+    embedding_model):
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    # define retriever object
+    retriever = vectordb.as_retriever(search_type="similarity", search_kwargs={"k": top_p})
 
-        response += token
-        yield response
+    # initialse chatbot llm
+    llm = HuggingFaceHub(
+        repo_id=MISTRAL_MODEL1,
+        token=os.environ["HUGGINGFACEHUB_API_TOKEN"], #huggingfacehub_api_token=SECRET_TOKEN_HF,
+        model_kwargs={"temperature": temperature, "max_new_tokens": max_tokens}
+        )
+    # create a RAG pipeline
+    qa_chain = RetrievalQA.from_chain_type(llm=llm, retriever=retriever)
+    #generate results
+    result = qa_chain.invoke(query)
+    
+    yield result['result']
 
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+        #gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.Slider(minimum=128, maximum=1024, value=512, step=128, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature"),
         gr.Slider(
             minimum=0.1,
             maximum=1.0,