Spaces:
Running
Running
File size: 5,263 Bytes
e04dd70 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 |
from langchain.chains import RetrievalQA
from langchain.prompts import PromptTemplate
from langchain.embeddings import OpenAIEmbeddings
# from langchain_community.embeddings import OpenAIEmbeddings
from langchain.vectorstores import Chroma
from src.utils import brian_knows_system_message
from uuid import uuid4
import chromadb
from chromadb.config import Settings
from chromadb.utils import embedding_functions
import sys
import os
import openai
import logging
sys.path.append("../..")
from dotenv import load_dotenv, find_dotenv
_ = load_dotenv(find_dotenv()) # read local .env file
openai.api_key = os.environ["OPENAI_API_KEY"]
class VectordbManager:
def __init__(
self,
knowledge_base_name: str,
) -> None:
self.knowledge_base_name = knowledge_base_name
self.vector_db = None
def load_vectordb(
self,
embedding_function=OpenAIEmbeddings(),
):
client = chromadb.HttpClient(
host="chroma.brianknows.org",
port="443",
ssl=True,
settings=Settings(allow_reset=True),
)
vectordb = Chroma(embedding_function=embedding_function, client=client)
self.vector_db = vectordb
def load_collection(self, embedding_function=OpenAIEmbeddings()):
client = chromadb.HttpClient(
host="chroma.brianknows.org",
port=443,
ssl=True,
settings=Settings(
allow_reset=True,
),
)
collection = client.get_collection(
self.knowledge_base_name,
embedding_function=embedding_functions.OpenAIEmbeddingFunction(
api_key=os.environ["OPENAI_API_KEY"]
),
)
return collection
def create_vector_db(self, splits: list, knowledge_base_name: str):
logging.info("create_vector_db")
embedding_fn = OpenAIEmbeddings()
try:
client = chromadb.HttpClient(
host="chroma.brianknows.org",
port=443,
ssl=True,
settings=Settings(
allow_reset=True,
),
)
collection = client.get_or_create_collection(
knowledge_base_name,
embedding_function=embedding_functions.OpenAIEmbeddingFunction(
api_key=os.environ["OPENAI_API_KEY"]
),
)
ids = []
metadatas = []
documents = []
for split in splits:
ids.append(str(uuid4()))
metadatas.append(split.metadata)
documents.append(split.page_content)
collection.add(documents=documents, ids=ids, metadatas=metadatas)
vector_db = Chroma.from_documents(
documents=splits, embedding=embedding_fn, client=client
)
self.vector_db = vector_db
except Exception as e:
logging.error(f"error in creating db: {str(e)}")
def add_splits_to_existing_vectordb(
self,
splits: list,
):
for split in splits:
try:
self.vector_db.add_documents([split])
print("document loaded!")
except Exception as e:
print(f"Error with doc : {split}")
print(e)
def retrieve_docs_from_query(self, query: str, k=2, fetch_k=3) -> list:
"""
query : Text to look up documents similar to.
k : Number of Documents to return. Defaults to 4.
fetch_k : Number of Documents to fetch to pass to MMR algorithm.
lambda_mult : Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5.
"""
retrieved_docs = self.vector_db.max_marginal_relevance_search(
query, k=k, fetch_k=fetch_k
)
return retrieved_docs
def retrieve_qa(
self,
llm,
query: str,
score_threshold: float = 0.65,
system_message=brian_knows_system_message,
):
"""return llm answer based on docs"""
# Build prompt
template = """You are a Web3 assistant. Use the following pieces of context to answer the question at \
the end. If you don't know the answer, just say: "I don't know". Don't try to make up an \
answer! Provide a always a detailed and comprehensive response. """
fixed_template = """ {context}
Question: {question}
Detailed Answer:"""
template = system_message + fixed_template
QA_CHAIN_PROMPT = PromptTemplate.from_template(template)
# Run chain
qa_chain = RetrievalQA.from_chain_type(
llm,
retriever=self.vector_db.as_retriever(
search_type="similarity_score_threshold",
search_kwargs={"score_threshold": score_threshold},
),
return_source_documents=True,
chain_type_kwargs={"prompt": QA_CHAIN_PROMPT},
# reduce_k_below_max_tokens=True,
)
result = qa_chain({"query": query})
return result
|