Spaces:
Sleeping
Sleeping
Commit
·
952eb35
1
Parent(s):
5aee298
Upload TXT and PDF files
Browse files
app.py
CHANGED
@@ -1,5 +1,6 @@
|
|
1 |
import os
|
2 |
import streamlit as st
|
|
|
3 |
from io import StringIO
|
4 |
|
5 |
#for textfiles
|
@@ -18,12 +19,42 @@ from langchain.document_loaders import UnstructuredPDFLoader
|
|
18 |
|
19 |
os.environ["HUGGINGFACEHUB_API_TOKEN"] = st.secrets["hf_api_key"]
|
20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
st.title('Document Q&A - Ask anything in your Document')
|
|
|
22 |
st.sidebar.subheader('Upload document')
|
23 |
uploaded_file = st.sidebar.file_uploader("Upload File",type=['txt','pdf'])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
with st.sidebar.expander('File'):
|
25 |
-
if(uploaded_file):
|
26 |
-
|
|
|
|
|
|
|
27 |
|
28 |
|
29 |
|
@@ -42,12 +73,6 @@ if(uploaded_file):
|
|
42 |
loader = TextLoader(uploaded_file.name)
|
43 |
documents = loader.load()
|
44 |
|
45 |
-
# Document Loader
|
46 |
-
#loader = TextLoader('./KS-all-info_rev1.txt')
|
47 |
-
|
48 |
-
# loader = TextLoader(os.path.join("./", uploaded_file.name))
|
49 |
-
|
50 |
-
|
51 |
# import textwrap
|
52 |
# def wrap_text_preserve_newlines(text, width=110):
|
53 |
# # Split the input text into lines based on newline characters
|
@@ -78,8 +103,6 @@ if(uploaded_file):
|
|
78 |
docs = db.similarity_search(query)
|
79 |
answer = chain.run(input_documents=docs, question=query)
|
80 |
|
81 |
-
|
82 |
-
|
83 |
st.subheader('Answer')
|
84 |
st.write(answer)
|
85 |
|
|
|
1 |
import os
|
2 |
import streamlit as st
|
3 |
+
from pathlib import Path
|
4 |
from io import StringIO
|
5 |
|
6 |
#for textfiles
|
|
|
19 |
|
20 |
os.environ["HUGGINGFACEHUB_API_TOKEN"] = st.secrets["hf_api_key"]
|
21 |
|
22 |
+
def pdf_file(filename):
|
23 |
+
st.subheader('Uploaded PDF File:')
|
24 |
+
st.write(filename)
|
25 |
+
|
26 |
+
def text_file(filename):
|
27 |
+
st.subheader('Uploaded Text File:')
|
28 |
+
st.write(filename)
|
29 |
+
|
30 |
+
# loader = TextLoader(filename)
|
31 |
+
# documents = loader.load()
|
32 |
+
|
33 |
+
# # Text Splitter
|
34 |
+
# text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=10)
|
35 |
+
# docs = text_splitter.split_documents(documents)
|
36 |
+
|
37 |
+
# db = FAISS.from_documents(docs, embeddings)
|
38 |
+
|
39 |
+
# chain = load_qa_chain(llm2, chain_type="stuff")
|
40 |
+
|
41 |
st.title('Document Q&A - Ask anything in your Document')
|
42 |
+
st.subheader('This application can be used to upload text(.txt) and PDF(.pdf) files and ask questions about their contents.')
|
43 |
st.sidebar.subheader('Upload document')
|
44 |
uploaded_file = st.sidebar.file_uploader("Upload File",type=['txt','pdf'])
|
45 |
+
|
46 |
+
if Path(uploaded_file.name).suffix == '.txt':
|
47 |
+
text_file(uploaded_file.name)
|
48 |
+
|
49 |
+
if Path(uploaded_file.name).suffix == '.pdf':
|
50 |
+
pdf_file(uploaded_file.name)
|
51 |
+
|
52 |
with st.sidebar.expander('File'):
|
53 |
+
if (uploaded_file):
|
54 |
+
st.info(uploaded_file.name)
|
55 |
+
if os.path.exists('/content/'):
|
56 |
+
st.info(os.listdir('/content/'))
|
57 |
+
|
58 |
|
59 |
|
60 |
|
|
|
73 |
loader = TextLoader(uploaded_file.name)
|
74 |
documents = loader.load()
|
75 |
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
# import textwrap
|
77 |
# def wrap_text_preserve_newlines(text, width=110):
|
78 |
# # Split the input text into lines based on newline characters
|
|
|
103 |
docs = db.similarity_search(query)
|
104 |
answer = chain.run(input_documents=docs, question=query)
|
105 |
|
|
|
|
|
106 |
st.subheader('Answer')
|
107 |
st.write(answer)
|
108 |
|