Spaces:
Running
Running
| # fix Error: module 'langchain' has no attribute 'verbose' | |
| import langchain | |
| import streamlit as st | |
| from langchain.callbacks import get_openai_callback | |
| from langchain.chains import ConversationalRetrievalChain | |
| from langchain.chat_models import ChatOpenAI | |
| from langchain.prompts.prompt import PromptTemplate | |
| langchain.verbose = False | |
| class Chatbot: | |
| def __init__(self, model_name, temperature, vectors): | |
| self.model_name = model_name | |
| self.temperature = temperature | |
| self.vectors = vectors | |
| qa_template = """ | |
| You are a helpful AI assistant named Robby. The user gives you a file its content is represented by the following pieces of context, use them to answer the question at the end. | |
| If you don't know the answer, just say you don't know. Do NOT try to make up an answer. | |
| If the question is not related to the context, politely respond that you are tuned to only answer questions that are related to the context. | |
| Use as much detail as possible when responding. | |
| context: {context} | |
| ========= | |
| question: {question} | |
| ====== | |
| """ | |
| QA_PROMPT = PromptTemplate( | |
| template=qa_template, input_variables=["context", "question"] | |
| ) | |
| def conversational_chat(self, query): | |
| """ | |
| Start a conversational chat with a model via Langchain | |
| """ | |
| # llm = ChatOpenAI(model_name=self.model_name, temperature=self.temperature) | |
| from modules.llm import ChatGLM | |
| llm = ChatGLM() | |
| retriever = self.vectors.as_retriever() | |
| chain = ConversationalRetrievalChain.from_llm( | |
| llm=llm, | |
| retriever=retriever, | |
| verbose=True, | |
| return_source_documents=True, | |
| max_tokens_limit=4097, | |
| combine_docs_chain_kwargs={"prompt": self.QA_PROMPT}, | |
| ) | |
| chain_input = {"question": query, "chat_history": st.session_state["history"]} | |
| with get_openai_callback() as cb: | |
| result = chain(chain_input) | |
| st.session_state["history"].append((query, result["answer"])) | |
| # count_tokens_chain(chain, chain_input) | |
| st.write( | |
| f"###### Tokens used in this conversation : {cb.total_tokens} tokens" | |
| ) | |
| return result["answer"] | |
| def count_tokens_chain(chain, query): | |
| with get_openai_callback() as cb: | |
| result = chain(query) | |
| st.write(f"###### Tokens used in this conversation : {cb.total_tokens} tokens") | |
| return result | |