-
Notifications
You must be signed in to change notification settings - Fork 0
/
embedding_ui.py
49 lines (30 loc) · 1.25 KB
/
embedding_ui.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
import streamlit as st
import os
import database_functions as dbf
import chromadb
import uuid
import datetime
st.title('Embedding UI')
# Upload Documents
uploaded_files = st.file_uploader("Choose a file", accept_multiple_files=True)
st.write(uploaded_files)
from langchain_community.document_loaders import TextLoader
from langchain.text_splitter import SentenceTransformersTokenTextSplitter
file_info = uploaded_files[0].read().decode("utf-8")
from langchain.text_splitter import RecursiveCharacterTextSplitter
text_splitter = RecursiveCharacterTextSplitter(
# Set a really small chunk size, just to show.
chunk_size=400,
chunk_overlap=40,
length_function=len,
is_separator_regex=False,
)
texts = text_splitter.create_documents([file_info])
st.write(texts)
embed_button = st.button("Embed")
if embed_button:
client = chromadb.PersistentClient(path="/vector_databases/test")
embedding_texts = [texts[i].page_content for i in range(len(texts))]
metadata = [{"source": uploaded_files[0].name, "date": str(datetime.datetime.now().date()), "source_order": i} for i in range(len(texts))]
ids = [str(uuid.uuid4()) for i in range(len(texts))]
dbf.add_embedding(client, "test_collection", embedding_texts, metadata, ids)