Commit
·
eb51c13
1
Parent(s):
74acb04
Update app.py
Browse files
app.py
CHANGED
@@ -1,13 +1,23 @@
|
|
1 |
import streamlit as st
|
2 |
from transformers import pipeline
|
|
|
3 |
|
4 |
st.title('Hugging Face BERT Summarizer')
|
5 |
uploaded_file = st.file_uploader("Choose a .txt file", type="txt")
|
6 |
|
7 |
if uploaded_file is not None:
|
8 |
user_input = uploaded_file.read().decode('utf-8')
|
|
|
9 |
if st.button('Summarize'):
|
10 |
summarizer = pipeline('summarization')
|
11 |
-
|
12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
st.text_area('Summarized Text', summarized_text, height=200)
|
|
|
1 |
import streamlit as st
|
2 |
from transformers import pipeline
|
3 |
+
import textwrap
|
4 |
|
5 |
st.title('Hugging Face BERT Summarizer')
|
6 |
uploaded_file = st.file_uploader("Choose a .txt file", type="txt")
|
7 |
|
8 |
if uploaded_file is not None:
|
9 |
user_input = uploaded_file.read().decode('utf-8')
|
10 |
+
|
11 |
if st.button('Summarize'):
|
12 |
summarizer = pipeline('summarization')
|
13 |
+
summarized_text = ""
|
14 |
+
|
15 |
+
# Split the text into chunks of approximately 500 words each
|
16 |
+
chunks = textwrap.wrap(user_input, 500)
|
17 |
+
|
18 |
+
# Summarize each chunk
|
19 |
+
for chunk in chunks:
|
20 |
+
summarized = summarizer(chunk, max_length=130, min_length=30, do_sample=False)
|
21 |
+
summarized_text += summarized[0]['summary_text'] + " "
|
22 |
+
|
23 |
st.text_area('Summarized Text', summarized_text, height=200)
|