-
Notifications
You must be signed in to change notification settings - Fork 0
/
create_database.py
63 lines (45 loc) · 1.41 KB
/
create_database.py
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
import os
import shutil
from langchain.document_loaders import DirectoryLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_openai import OpenAIEmbeddings
from langchain_community.vectorstores import Chroma
CHROMA_PATH = "chroma"
DATA_PATH = "data/books"
def main():
generate_data_store()
def generate_data_store():
documents = load_documents()
chunks = split_text(documents)
save_to_chroma(chunks)
def load_documents():
loader = DirectoryLoader(DATA_PATH, glob="*.md")
documents = loader.load()
return documents
def split_text(documents):
text_splitter = RecursiveCharacterTextSplitter(
chunk_size=1000,
chunk_overlap=100,
length_function=len,
add_start_index=True,
)
chunks = text_splitter.split_documents(documents)
print(f"Split {len(documents)} documents into {len(chunks)} chunks.")
example = chunks[2]
print(f"Example chunk: {example.page_content}")
print(example.metadata)
return chunks
def save_to_chroma(chunks) -> None:
# Clear database first
if os.path.exists(CHROMA_PATH):
shutil.rmtree(CHROMA_PATH)
# Create new database
Chroma.from_documents(
chunks,
OpenAIEmbeddings(),
collection_name="books",
persist_directory=CHROMA_PATH,
)
print(f"Saved {len(chunks)} chunks to {CHROMA_PATH}")
if __name__ == "__main__":
main()