app-original.py

import os
import requests
from langchain.agents import Tool
from langchain.tools import BaseTool
from langchain.agents import load_tools
from langchain.memory import ConversationBufferMemory
from langchain.memory import ConversationBufferWindowMemory
#from langchain.chat_models import ChatOpenAI
from langchain.utilities import GoogleSearchAPIWrapper
from langchain.utilities import GoogleSerperAPIWrapper
from langchain.agents import initialize_agent
import gradio as gr 
from langchain.chains.question_answering import load_qa_chain
from langchain import PromptTemplate, LLMChain
from langchain import HuggingFaceHub
from pathlib import Path
from time import sleep
from langchain.agents import AgentType
#from langchain.llms import OpenAI

from langchain.agents import AgentOutputParser
from langchain.agents.conversational_chat.prompt import FORMAT_INSTRUCTIONS
from langchain.output_parsers.json import parse_json_markdown
from langchain.schema import AgentAction, AgentFinish

import os
import random
import string
from dotenv import load_dotenv
load_dotenv()

from langchain.agents import ZeroShotAgent, Tool, AgentExecutor
from langchain import OpenAI, SerpAPIWrapper, LLMChain, LLMMathChain
     

OPENAI_API_KEY =os.getenv("OPENAI_API_KEY")
GOOGLE_API_KEY =os.getenv("GOOGLE_API_KEY")
GOOGLE_CSE_ID =os.getenv("GOOGLE_CSE_ID")
#HUGGINGFACEHUB_API_TOKEN = os.getenv('HUGGINGFACEHUB_API_TOKEN')
#repo_id = os.getenv('repo_id')
HUGGINGFACEHUB_API_TOKEN = os.environ.get('HUGGINGFACEHUB_API_TOKEN')
SERPAPI_API_KEY=os.environ.get('SERPAPI_API_KEY')
repo_id = os.environ.get('repo_id')

template = """Question: {question}

Answer: Let's think step by step."""

prompt_lora = PromptTemplate(template=template, input_variables=["question"])

llm=HuggingFaceHub(repo_id=repo_id)

lora_chain = LLMChain(prompt=prompt_lora,llm = llm)

question = "How many fish can live in a ocean that is as big as North America?"

print(lora_chain.run(question))

# This is an LLMChain to write a synopsis given a title of a play.

template = """You are a engineer. Given the title, it is your job to write a synopsis for that title.

Title: {title}
Engineer: This is a synopsis for the above title:"""

prompt_template = PromptTemplate(input_variables=["title"], 
                                 template=template)

synopsis_chain = LLMChain(llm=llm, 
                          prompt=prompt_template)

template = """You are a MBA from Harvard. Given the synopsis, it is your job to write a review for that synopsis.

Synopsis:
{synopsis}
Review from a Harvard MBA of above synopsis:"""

prompt_template = PromptTemplate(input_variables=["synopsis"], template=template)

review_chain = LLMChain(llm=llm, prompt=prompt_template)

from langchain.chains import SimpleSequentialChain
overall_chain = SimpleSequentialChain(chains=[synopsis_chain, review_chain], verbose=True)

overall_chain.run("How big is London Bridge?")


search = SerpAPIWrapper()
llm_math_chain = LLMMathChain(llm=llm, verbose=True)

tools = [
    Tool(
        name = "Search",
        func=search.run,
        description="useful for when you need to answer questions about current events"
    ),
    Tool(
        name="Calculator",
        func=llm_math_chain.run,
        description="useful for when you need to answer questions about Divide, Multiply, Add and Subtract"
    )
]


prefix = """Answer the following questions as best you can. Think through step by step.
You have access to the following tools:"""
suffix = """Begin! Remember to think step by step

Question: {input}
{agent_scratchpad}"""

prompt = ZeroShotAgent.create_prompt(
    tools, 
    prefix=prefix, 
    suffix=suffix, 
    input_variables=["input", "agent_scratchpad"]
)

agent_chain = LLMChain(llm=llm, prompt=prompt)

tool_names = [tool.name for tool in tools]
agent = ZeroShotAgent(llm_chain=agent_chain, 
                      allowed_tools=tool_names)

agent_executor = AgentExecutor.from_agent_and_tools(agent=agent, 
                                                    tools=tools, 
                                                    verbose=True,
                                                    max_iterations=3)

agent.save("custom_agent.json")

result=agent_executor.run("How many people live in canada as of 2023?")

print("Result: "+str(result))