Update app.py

app.py

@@ -1,64 +1,76 @@
+from langchain_core.prompts import ChatPromptTemplate
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.output_parsers import StrOutputParser
+from langchain_chroma import Chroma
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain_community.document_loaders import WebBaseLoader
+from langchain_openai import AzureChatOpenAI
 import gradio as gr
-from huggingface_hub import InferenceClient
 
-"""
-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")
 
+llm = AzureChatOpenAI(
+            openai_api_type="azure",
+            openai_api_version='2024-05-01-preview',
+            openai_api_key='7a8f58dd922e4c78b1de2b660ebe61d6',
+            azure_endpoint='https://mlsdaiinstance.openai.azure.com/',
+            deployment_name="gpt-4o",
+            temperature=0.5
+        )
 
-def respond(
-    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]})
+# loader = PyPDFDirectoryLoader("data")
+loader = WebBaseLoader(
+    web_paths=("https://vyomastra.in/index.html",
+               "https://vyomastra.in/about_us.html",
+               "https://vyomastra.in/solutions.html",
+                )
+)
 
-    messages.append({"role": "user", "content": message})
+text = loader.load()
+text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=200)
+docs = text_splitter.split_documents(text)
 
-    response = ""
+vectorstore = Chroma.from_documents(
+    documents=docs,
+    collection_name="embeds",
+    embedding=HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2"),
+)
 
-    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
+retriever = vectorstore.as_retriever()
 
-        response += token
-        yield response
 
+rag_template = """You are a conversational question answering AI assistant named Astra.
+You are created by AI developers from Vyomastra.
+Your abilities: logical reasoning, complex mathematics computing, coding knowledge, common general knowledge from internet.
+Use your abilities and knowledge from the context mentioned below to answer the questions truthfully:
 
+{context}
+
+Question: {question}
+Answer:
 """
-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.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+
+rag_prompt = ChatPromptTemplate.from_template(rag_template)
+rag_chain = (
+    {"context": retriever, "question": RunnablePassthrough()}
+    | rag_prompt
+    | llm
+    | StrOutputParser()
 )
 
+# Make the questions dynamic using a chat interface. Let's use gradio for this.
+def process_question(user_question):
+    response = rag_chain.invoke(user_question)
+    return response
+
+
+# Setup the Gradio interface
+iface = gr.Interface(fn=process_question,
+                     inputs=gr.Textbox(lines=2, placeholder="Type your question here..."),
+                     outputs=gr.Textbox(),
+                     title="Website Knowledge Chat App",
+                     description="Ask any question about your document, and get an answer along with the response time.")
 
-if __name__ == "__main__":
-    demo.launch()
+# Launch the interface
+iface.launch()