File size: 3,488 Bytes
d73d493
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
from aimakerspace.text_utils import TextFileLoader, CharacterTextSplitter
from aimakerspace.vectordatabase import VectorDatabase
import asyncio
import os
import openai
from getpass import getpass
from aimakerspace.openai_utils.prompts import (
    UserRolePrompt,
    SystemRolePrompt,
    AssistantRolePrompt,
)

from aimakerspace.openai_utils.chatmodel import ChatOpenAI
import chainlit as cl  # importing chainlit for our app
from chainlit.prompt import Prompt, PromptMessage  # importing prompt tools




text_loader = TextFileLoader("data/KingLear.txt")
documents = text_loader.load_documents()
len(documents)

print(documents[0][:600])


text_splitter = CharacterTextSplitter()
split_documents = text_splitter.split_texts(documents)
len(split_documents)

split_documents[0:1]

vector_db = VectorDatabase()
vector_db = asyncio.run(vector_db.abuild_from_list(split_documents))

#vector_db.search_by_text("Your servant Kent. Where is your servant Caius?", k=3)


chat_openai = ChatOpenAI()
user_prompt_template = "{content}"
user_role_prompt = UserRolePrompt(user_prompt_template)
system_prompt_template = (
    "You are an expert in {expertise}, you always answer in a kind way."
)
system_role_prompt = SystemRolePrompt(system_prompt_template)

messages = [
    user_role_prompt.create_message(
        content="What is the best way to write a loop?"
    ),
    system_role_prompt.create_message(expertise="Python"),
]




#response = chat_openai.run(messages)

#print(response)

RAQA_PROMPT_TEMPLATE = """
Use the provided context to answer the user's query. 

You may not answer the user's query unless there is specific context in the following text.

If you do not know the answer, or cannot answer, please respond with "I don't know".

Context:
{context}
"""

raqa_prompt = SystemRolePrompt(RAQA_PROMPT_TEMPLATE)


USER_PROMPT_TEMPLATE = """
User Query:
{user_query}
"""

user_prompt = UserRolePrompt(USER_PROMPT_TEMPLATE)

class RetrievalAugmentedQAPipeline:
    def __init__(self, llm: ChatOpenAI(), vector_db_retriever: VectorDatabase) -> None:
        self.llm = llm
        self.vector_db_retriever = vector_db_retriever

    def run_pipeline(self, user_query: str) -> str:
        context_list = self.vector_db_retriever.search_by_text(user_query, k=4)
        
        context_prompt = ""
        for context in context_list:
            context_prompt += context[0] + "\n"

        formatted_system_prompt = raqa_prompt.create_message(context=context_prompt)

        formatted_user_prompt = user_prompt.create_message(user_query=user_query)
        
        return self.llm.run([formatted_system_prompt, formatted_user_prompt])




    
retrieval_augmented_qa_pipeline = RetrievalAugmentedQAPipeline(
    vector_db_retriever=vector_db,
    llm=chat_openai
)
    


  
#print(retrieval_augmented_qa_pipeline.run_pipeline("Who is King Lear?"))



@cl.on_chat_start  # marks a function that will be executed at the start of a user session
async def start_chat():
    settings = {
        "model": "gpt-3.5-turbo",
        "temperature": 0,
        "max_tokens": 500,
        "top_p": 1,
        "frequency_penalty": 0,
        "presence_penalty": 0,
    }

    cl.user_session.set("settings", settings)




@cl.on_message  # marks a function that should be run each time the chatbot receives a message from a user
async def main(message: cl.Message):
    
    await cl.Message(content=retrieval_augmented_qa_pipeline.run_pipeline(message.content), elements=[]).send()