File size: 2,801 Bytes
48c4cb2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
!pip install -q -U numpy matplotlib plotly pandas scipy scikit-learn openai python-dotenv

from aimakerspace.text_utils import TextFileLoader, CharacterTextSplitter
from aimakerspace.vectordatabase import VectorDatabase
import asyncio

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

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

import os
import openai
from getpass import getpass

openai.api_key = getpass("OpenAI API Key: ")
os.environ["OPENAI_API_KEY"] = openai.api_key

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

import sys
sys.path.append('/home/rlpeter70/LLMO-Cohort-3/Week 1/Thursday - Retrieval Augmented Generation QA Application /aimakerspace/openai_utils')

from aimakerspace.openai_utils.prompts import (
    UserRolePrompt,
    SystemRolePrompt,
    AssistantRolePrompt,
)

from aimakerspace.openai_utils.chatmodel import ChatOpenAI

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)

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
)