mirror of
https://github.com/langchain-ai/streamlit-agent.git
synced 2026-07-01 09:25:05 -04:00
Joshua Carroll
80 lines
3.1 KiB
Python
80 lines
3.1 KiB
Python
from langchain.chains import ConversationChain
|
|
from langchain.memory import ConversationBufferMemory
|
|
from langchain_community.chat_message_histories import StreamlitChatMessageHistory
|
|
from langchain_core.runnables import RunnableConfig
|
|
from langchain_core.tracers import LangChainTracer
|
|
from langchain_core.tracers.run_collector import RunCollectorCallbackHandler
|
|
from langchain_openai import OpenAI
|
|
from langsmith import Client
|
|
import streamlit as st
|
|
from streamlit_feedback import streamlit_feedback
|
|
import time
|
|
|
|
st.set_page_config(page_title="LangChain: Simple feedback", page_icon="🦜")
|
|
st.title("🦜 LangChain: Simple feedback")
|
|
|
|
openai_api_key = st.secrets.get("OPENAI_API_KEY")
|
|
if not openai_api_key:
|
|
openai_api_key = st.sidebar.text_input("OpenAI API Key", type="password")
|
|
langchain_api_key = st.secrets.get("LANGCHAIN_API_KEY")
|
|
if not langchain_api_key:
|
|
langchain_api_key = st.sidebar.text_input("LangChain API Key", type="password")
|
|
project = st.sidebar.text_input("LangSmith Project", value="default")
|
|
if not langchain_api_key or not openai_api_key:
|
|
st.warning("Please add an OpenAI API Key and LangChain API Key to continue")
|
|
st.stop()
|
|
|
|
# Customize if needed
|
|
langchain_endpoint = "https://api.smith.langchain.com"
|
|
client = Client(api_url=langchain_endpoint, api_key=langchain_api_key)
|
|
ls_tracer = LangChainTracer(project_name=project, client=client)
|
|
run_collector = RunCollectorCallbackHandler()
|
|
cfg = RunnableConfig()
|
|
cfg["callbacks"] = [ls_tracer, run_collector]
|
|
|
|
msgs = StreamlitChatMessageHistory()
|
|
memory = ConversationBufferMemory(chat_memory=msgs)
|
|
llm_chain = ConversationChain(llm=OpenAI(openai_api_key=openai_api_key), memory=memory)
|
|
|
|
reset_history = st.sidebar.button("Reset chat history")
|
|
if len(msgs.messages) == 0 or reset_history:
|
|
msgs.clear()
|
|
msgs.add_ai_message("How can I help you?")
|
|
st.session_state["last_run"] = None
|
|
|
|
avatars = {"human": "user", "ai": "assistant"}
|
|
for msg in msgs.messages:
|
|
st.chat_message(avatars[msg.type]).write(msg.content)
|
|
|
|
if input := st.chat_input(placeholder="Tell me a joke about a shark?"):
|
|
st.chat_message("user").write(input)
|
|
with st.chat_message("assistant"):
|
|
response = llm_chain.invoke(input, cfg)
|
|
st.write(response["response"])
|
|
st.session_state.last_run = run_collector.traced_runs[0].id
|
|
|
|
|
|
@st.cache_data(ttl="2h", show_spinner=False)
|
|
def get_run_url(run_id):
|
|
time.sleep(1)
|
|
return client.read_run(run_id).url
|
|
|
|
|
|
if st.session_state.get("last_run"):
|
|
run_url = get_run_url(st.session_state.last_run)
|
|
st.sidebar.markdown(f"[Latest Trace: 🛠️]({run_url})")
|
|
feedback = streamlit_feedback(
|
|
feedback_type="faces",
|
|
optional_text_label="[Optional] Please provide an explanation",
|
|
key=f"feedback_{st.session_state.last_run}",
|
|
)
|
|
if feedback:
|
|
scores = {"😀": 1, "🙂": 0.75, "😐": 0.5, "🙁": 0.25, "😞": 0}
|
|
client.create_feedback(
|
|
st.session_state.last_run,
|
|
feedback["type"],
|
|
score=scores[feedback["score"]],
|
|
comment=feedback.get("text", None),
|
|
)
|
|
st.toast("Feedback recorded!", icon="📝")
|