Spaces:
Sleeping
Sleeping
first
Browse files
.env
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
OPENAI_API_KEY=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
|
app.py
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
from PyPDF2 import PdfReader
|
3 |
+
from streamlit_extras.add_vertical_space import add_vertical_space
|
4 |
+
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
5 |
+
from langchain.embeddings.openai import OpenAIEmbeddings
|
6 |
+
from langchain.vectorstores import faiss
|
7 |
+
|
8 |
+
with st.sidebar:
|
9 |
+
st.title("File Research using LLM")
|
10 |
+
st.markdown(''' Upload your file and ask questions and do Research''')
|
11 |
+
add_vertical_space(5)
|
12 |
+
pdf=st.file_uploader('Upload your file (PDF)', type='pdf')
|
13 |
+
if pdf is not None:
|
14 |
+
pdf_reader=PdfReader(pdf)
|
15 |
+
text=""
|
16 |
+
for page in pdf_reader.pages:
|
17 |
+
text+=page.extract_text()
|
18 |
+
text_splitter=RecursiveCharacterTextSplitter(
|
19 |
+
chunk_size=1000,
|
20 |
+
chunk_overlap=200,
|
21 |
+
length_function=len
|
22 |
+
)
|
23 |
+
chunks=text_splitter.split_text(text)
|
24 |
+
embeddings=OpenAIEmbeddings()
|
25 |
+
vectorstore=faiss.FAISS.from_texts(chunks, embedding=embeddings)
|
26 |
+
|
27 |
+
|
28 |
+
|
29 |
+
st.write(chunks)
|
30 |
+
st.write('Made by ALOK')
|
31 |
+
|
32 |
+
def main():
|
33 |
+
st.header('Talk to your file')
|
34 |
+
|
35 |
+
if __name__=='__main__':
|
36 |
+
main()
|
requirements.txt
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
python==3.11.5
|
2 |
+
langchain
|
3 |
+
PyPDF2
|
4 |
+
streamlit
|
5 |
+
python-dotenv
|
6 |
+
faiss-cpu
|
7 |
+
streamlit-extras
|
test.py
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
print("hi")
|