Create app

app

@@ -0,0 +1,100 @@
+from langchain import PromptTemplate
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain_community.llms import CTransformers
+from langchain.chains import RetrievalQA
+import gradio as gr
+from huggingface_hub import hf_hub_download
+
+DB_FAISS_PATH = "vectorstores/db_faiss"
+
+def load_llm():
+    """
+    Load the LLaMA model for the language model.
+    """
+    model_name = 'TheBloke/Llama-2-7B-Chat-GGML'
+    model_path = hf_hub_download(repo_id=model_name, filename='llama-2-7b-chat.ggmlv3.q8_0.bin', cache_dir='./models')
+    llm = CTransformers(
+        model=model_path,
+        model_type="llama",
+        max_new_tokens=512,
+        temperature=0.5
+    )
+    return llm
+
+def set_custom_prompt():
+    """
+    Define a custom prompt template for the QA model.
+    """
+    custom_prompt_template = """Use the following pieces of information to answer the user's question.
+If you don't know the answer, just say that you don't know, don't try to make up an answer.
+
+Context: {context}
+Question: {question}
+
+only return the helpful answer below and nothing else.
+Helpful answer:
+"""
+    prompt = PromptTemplate(template=custom_prompt_template, input_variables=['context', 'question'])
+    return prompt
+
+def retrieval_QA_chain(llm, prompt, db):
+    """
+    Create a RetrievalQA chain with the specified LLM, prompt, and vector store.
+    """
+    qachain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=db.as_retriever(search_kwargs={'k': 2}),
+        return_source_documents=True,
+        chain_type_kwargs={'prompt': prompt}
+    )
+    return qachain
+
+def qa_bot():
+    """
+    Initialize the QA bot with embeddings, vector store, LLM, and prompt.
+    """
+    embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-miniLM-L6-V2', model_kwargs={'device': 'cpu'})
+    db = FAISS.load_local(DB_FAISS_PATH, embeddings, allow_dangerous_deserialization=True)
+    llm = load_llm()
+    qa_prompt = set_custom_prompt()
+    qa = retrieval_QA_chain(llm, qa_prompt, db)
+    return qa
+
+bot = qa_bot()
+
+def chatbot_response(message, history):
+    """
+    Generate a response from the chatbot based on the user input and conversation history.
+    """
+    try:
+        response = bot({'query': message})
+        answer = response["result"]
+        sources = response["source_documents"]
+        if sources:
+            answer += f"\nSources: {sources}"
+        else:
+            answer += "\nNo sources found"
+        history.append((message, answer))
+    except Exception as e:
+        history.append((message, f"An error occurred: {str(e)}"))
+    return history, history
+
+# Set up the Gradio interface
+demo = gr.Interface(
+    fn=chatbot_response,
+    inputs=[
+        gr.Textbox(label="User Input"),
+        gr.State(value=[], label="Conversation History")
+    ],
+    outputs=[
+        gr.Chatbot(label="Chatbot Response"),
+        gr.State()
+    ],
+    title="AdvocateAI",
+    description="Ask questions about AI rights and get informed, passionate answers."
+)
+
+if __name__ == "__main__":
+    demo.launch()