innofinderai / mixtral_agent.py
isayahc's picture
found a way to collect sources from tools
519bb78
raw
history blame
6.21 kB
# LangChain supports many other chat models. Here, we're using Ollama
from langchain_community.chat_models import ChatOllama
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain.tools.retriever import create_retriever_tool
from langchain_community.utilities import SerpAPIWrapper
from langchain.retrievers import ArxivRetriever
from langchain_core.tools import Tool
from langchain import hub
from langchain.agents import AgentExecutor, load_tools
from langchain.agents.format_scratchpad import format_log_to_str
from langchain.agents.output_parsers import (
ReActJsonSingleInputOutputParser,
)
from langchain.tools.render import render_text_description
import os
import dotenv
dotenv.load_dotenv()
OLLMA_BASE_URL = os.getenv("OLLMA_BASE_URL")
# supports many more optional parameters. Hover on your `ChatOllama(...)`
# class to view the latest available supported parameters
llm = ChatOllama(
model="mistral:instruct",
base_url= OLLMA_BASE_URL
)
prompt = ChatPromptTemplate.from_template("Tell me a short joke about {topic}")
# using LangChain Expressive Language chain syntax
# learn more about the LCEL on
# https://python.langchain.com/docs/expression_language/why
chain = prompt | llm | StrOutputParser()
# for brevity, response is printed in terminal
# You can use LangServe to deploy your application for
# production
print(chain.invoke({"topic": "Space travel"}))
retriever = ArxivRetriever(load_max_docs=2)
# Import things that are needed generically
from langchain.pydantic_v1 import BaseModel, Field
from langchain.tools import BaseTool, StructuredTool, tool
global all_sources
# @tool
# def search(query: str) -> str:
# """Look up things online."""
# # return "LangChain"
# data = retriever.invoke(query)
# meta_data = [i.metadata for i in data]
# # meta_data += all_sources
# # all_sources += meta_data
# all_sources += meta_data
# # all_sources = []
# return meta_data
from typing import List, Dict
from datetime import datetime
def format_info_list(info_list: List[Dict[str, str]]) -> str:
"""
Format a list of dictionaries containing information into a single string.
Args:
info_list (List[Dict[str, str]]): A list of dictionaries containing information.
Returns:
str: A formatted string containing the information from the list.
"""
formatted_strings = []
for info_dict in info_list:
formatted_string = "|"
for key, value in info_dict.items():
if isinstance(value, datetime.date):
value = value.strftime('%Y-%m-%d')
formatted_string += f"'{key}': '{value}', "
formatted_string = formatted_string.rstrip(', ') + "|"
formatted_strings.append(formatted_string)
return '\n'.join(formatted_strings)
@tool
def search(query: str) -> str:
"""Look up things online."""
# return "LangChain"
global all_sources
data = retriever.invoke(query)
meta_data = [i.metadata for i in data]
# meta_data += all_sources
# all_sources += meta_data
all_sources += meta_data
# formatted_info = format_info(entry_id, published, title, authors)
# formatted_info = format_info_list(all_sources)
return meta_data.__str__()
# all_sources = []
# return meta_data
tools = [search]
# tools = [
# create_retriever_tool(
# retriever,
# "search arxiv's database for",
# "Use this to recomend the user a paper to read Unless stated please choose the most recent models",
# # "Searches and returns excerpts from the 2022 State of the Union.",
# ),
# Tool(
# name="SerpAPI",
# description="A low-cost Google Search API. Useful for when you need to answer questions about current events. Input should be a search query.",
# func=SerpAPIWrapper().run,
# )
# ]
prompt = hub.pull("hwchase17/react-json")
prompt = prompt.partial(
tools=render_text_description(tools),
tool_names=", ".join([t.name for t in tools]),
)
chat_model = llm
# define the agent
chat_model_with_stop = chat_model.bind(stop=["\nObservation"])
agent = (
{
"input": lambda x: x["input"],
"agent_scratchpad": lambda x: format_log_to_str(x["intermediate_steps"]),
}
| prompt
| chat_model_with_stop
| ReActJsonSingleInputOutputParser()
)
# instantiate AgentExecutor
agent_executor = AgentExecutor(
agent=agent,
tools=tools,
verbose=True,
handle_parsing_errors=True #prevents error
)
# agent_executor.invoke(
# {
# "input": "Who is the current holder of the speed skating world record on 500 meters? What is her current age raised to the 0.43 power?"
# }
# )
# agent_executor.invoke(
# {
# "input": "what are large language models and why are they so expensive to run?"
# }
# )
# agent_executor.invoke(
# {
# "input": "How to generate videos from images using state of the art macchine learning models"
# }
# )
# class AgentSample:
# def __init__(self, agent_executor_object,*args, **kwargs):
# self.agent_executor_object = agent_executor_object
# self.meta_data = []
# def sample_invokex
if __name__ == "__main__":
# global variable for collecting sources
all_sources = []
input = agent_executor.invoke(
{
"input": "How to generate videos from images using state of the art macchine learning models; Using the axriv retriever " +
"add the urls of the papers used in the final answer using the metadata from the retriever"
# f"Please prioritize the newest papers this is the current data {get_current_date()}"
}
)
x = 0
input_1 = agent_executor.invoke(
{
"input": "I am looking for a text to 3d model; Using the axriv retriever " +
"add the urls of the papers used in the final answer using the metadata from the retriever"
# f"Please prioritize the newest papers this is the current data {get_current_date()}"
}
)
x = 0