Spaces:
Runtime error
Runtime error
Upload 7 files
Browse files- .streamlit/config.toml +3 -0
- WebComponents/ArticleExtractor.py +26 -0
- app.py +16 -0
- packages.txt +4 -0
- pages/Extraction.py +55 -0
- pages/Final.py +16 -0
- requirements.txt +9 -0
.streamlit/config.toml
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[theme]
|
| 2 |
+
base = "dark"
|
| 3 |
+
font = "monospace"
|
WebComponents/ArticleExtractor.py
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import pytesseract
|
| 2 |
+
from PIL import Image
|
| 3 |
+
from io import BytesIO
|
| 4 |
+
#import easyocr
|
| 5 |
+
|
| 6 |
+
# Function to perform OCR with Tesseract
|
| 7 |
+
def ocr_with_tesseract(image):
|
| 8 |
+
# Perform OCR on the image using pytesseract
|
| 9 |
+
extracted_text = pytesseract.image_to_string(image)
|
| 10 |
+
return extracted_text
|
| 11 |
+
|
| 12 |
+
# Function to perform OCR with EasyOCR
|
| 13 |
+
'''
|
| 14 |
+
def ocr_with_easyocr(image):
|
| 15 |
+
# Convert the image to bytes for easyocr
|
| 16 |
+
image_bytes = image.tobytes()
|
| 17 |
+
|
| 18 |
+
# Perform OCR on the image using easyocr
|
| 19 |
+
reader = easyocr.Reader(['en']) # You can add other languages as needed
|
| 20 |
+
result = reader.readtext(image_bytes)
|
| 21 |
+
|
| 22 |
+
# Extract text from OCR result
|
| 23 |
+
extracted_text = [text[1] for text in result]
|
| 24 |
+
|
| 25 |
+
return extracted_text
|
| 26 |
+
'''
|
app.py
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import streamlit as st
|
| 2 |
+
from headline_gen.Control import ServerInit
|
| 3 |
+
import os
|
| 4 |
+
import nltk
|
| 5 |
+
|
| 6 |
+
def main():
|
| 7 |
+
#os.environ['JAVAHOME'] = 'usr/bin/java'
|
| 8 |
+
st.title("Abstract Page")
|
| 9 |
+
if "Server" not in st.session_state:
|
| 10 |
+
with st.spinner("Booting Server"):
|
| 11 |
+
st.session_state["Server"] = ServerInit("Start")
|
| 12 |
+
else:
|
| 13 |
+
if st.button("Upload Text"):
|
| 14 |
+
st.switch_page("pages/Extraction.py")
|
| 15 |
+
if __name__ == "__main__":
|
| 16 |
+
main()
|
packages.txt
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
tesseract-ocr
|
| 2 |
+
tesseract-ocr-por
|
| 3 |
+
default-jdk
|
| 4 |
+
default-jre
|
pages/Extraction.py
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import streamlit as st
|
| 2 |
+
from WebComponents import ArticleExtractor as ocr
|
| 3 |
+
|
| 4 |
+
from streamlit_cropper import st_cropper
|
| 5 |
+
from PIL import Image
|
| 6 |
+
|
| 7 |
+
st.set_page_config(layout="wide")
|
| 8 |
+
def main():
|
| 9 |
+
tab1, tab2 = st.tabs(['Manual', 'OCR'])
|
| 10 |
+
if 'Questions' not in st.session_state:
|
| 11 |
+
st.session_state['Questions'] = []
|
| 12 |
+
st.session_state['Flag'] = False
|
| 13 |
+
with tab1:
|
| 14 |
+
with st.form("Manual", clear_on_submit = True):
|
| 15 |
+
Question = st.text_area("Question")
|
| 16 |
+
st.write(Question)
|
| 17 |
+
col1, col2 = st.columns(2)
|
| 18 |
+
with col1:
|
| 19 |
+
if st.form_submit_button("Reset"):
|
| 20 |
+
k = 0
|
| 21 |
+
#st.session_state['Questions'].append({"QuestionID": ID, "Question": Question})
|
| 22 |
+
with col2:
|
| 23 |
+
if st.form_submit_button("Generate HL"):
|
| 24 |
+
st.session_state['Questions'].append(Question)
|
| 25 |
+
st.switch_page("pages/Final.py")
|
| 26 |
+
with tab2:
|
| 27 |
+
if "file_uploader_key" not in st.session_state:
|
| 28 |
+
st.session_state["file_uploader_key"] = 0
|
| 29 |
+
Image_File = st.file_uploader("Upload the Image", type=['png', 'jpg'], key=st.session_state["file_uploader_key"],)
|
| 30 |
+
if Image_File:
|
| 31 |
+
col1, col2 = st.columns(2)
|
| 32 |
+
with col1:
|
| 33 |
+
img = Image.open(Image_File)
|
| 34 |
+
cropped_img = st_cropper(img, realtime_update = True, box_color = "#FF0012", aspect_ratio = (5, 2))
|
| 35 |
+
|
| 36 |
+
with col2:
|
| 37 |
+
st.write("Preview")
|
| 38 |
+
_ = cropped_img.thumbnail((580, 580))
|
| 39 |
+
st.image(cropped_img)
|
| 40 |
+
#if st.button("Extract"):
|
| 41 |
+
ExText = ocr.ocr_with_tesseract(cropped_img)
|
| 42 |
+
with st.form("OCR", clear_on_submit = True):
|
| 43 |
+
Question = st.text_area("Question", value = ExText)
|
| 44 |
+
col1, col2 = st.columns(2)
|
| 45 |
+
with col1:
|
| 46 |
+
if st.form_submit_button("Reset"):
|
| 47 |
+
st.session_state["file_uploader_key"] += 1
|
| 48 |
+
st.rerun()
|
| 49 |
+
#st.session_state['Questions'].append({"Question": Question})
|
| 50 |
+
with col2:
|
| 51 |
+
if st.form_submit_button("Generate HL"):
|
| 52 |
+
st.session_state['Questions'].append(Question)
|
| 53 |
+
st.switch_page("pages/Final.py")
|
| 54 |
+
if __name__ == "__main__":
|
| 55 |
+
main()
|
pages/Final.py
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import streamlit as st
|
| 2 |
+
from headline_gen.Control import Generate
|
| 3 |
+
|
| 4 |
+
def main():
|
| 5 |
+
if 'Questions' in st.session_state:
|
| 6 |
+
with st.spinner("Generating Head Line!!!"):
|
| 7 |
+
#st.subheader(st.session_state['Questions'][-1])
|
| 8 |
+
if 'Server' in st.session_state:
|
| 9 |
+
st.success(Generate(st.session_state['Questions'][-1], st.session_state["Server"]))
|
| 10 |
+
else:
|
| 11 |
+
st.error("Server is not Booted")
|
| 12 |
+
else:
|
| 13 |
+
st.error("Please Navigate back to Extraction Page and Upload your Article")
|
| 14 |
+
|
| 15 |
+
if __name__ == "__main__":
|
| 16 |
+
main()
|
requirements.txt
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
streamlit-cropper
|
| 2 |
+
pytesseract
|
| 3 |
+
textacy
|
| 4 |
+
regex
|
| 5 |
+
nltk
|
| 6 |
+
scipy==1.12.0
|
| 7 |
+
gensim
|
| 8 |
+
networkx
|
| 9 |
+
headline-gen
|