Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -101,12 +101,11 @@
|
|
| 101 |
# if __name__ == "__main__":
|
| 102 |
# main()
|
| 103 |
|
| 104 |
-
|
| 105 |
import os
|
| 106 |
-
import PyPDF2
|
| 107 |
import logging
|
| 108 |
import math
|
| 109 |
import streamlit as st
|
|
|
|
| 110 |
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
|
| 111 |
from langchain_community.document_loaders import PDFMinerLoader
|
| 112 |
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
|
@@ -135,14 +134,14 @@ base_model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint)
|
|
| 135 |
# Helper Functions
|
| 136 |
|
| 137 |
def extract_text_from_pdf(file_path):
|
| 138 |
-
"""Extract text from a PDF using
|
| 139 |
try:
|
| 140 |
-
|
| 141 |
-
|
| 142 |
-
|
| 143 |
-
|
| 144 |
-
|
| 145 |
-
|
| 146 |
except Exception as e:
|
| 147 |
logging.error(f"Error reading PDF {file_path}: {e}")
|
| 148 |
return None
|
|
@@ -161,7 +160,7 @@ def data_ingestion():
|
|
| 161 |
file_path = os.path.join(uploaded_files_dir, filename)
|
| 162 |
logging.info(f"Processing file: {file_path}")
|
| 163 |
|
| 164 |
-
# Extract text using
|
| 165 |
text = extract_text_from_pdf(file_path)
|
| 166 |
|
| 167 |
if text:
|
|
|
|
| 101 |
# if __name__ == "__main__":
|
| 102 |
# main()
|
| 103 |
|
|
|
|
| 104 |
import os
|
|
|
|
| 105 |
import logging
|
| 106 |
import math
|
| 107 |
import streamlit as st
|
| 108 |
+
import fitz # PyMuPDF
|
| 109 |
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
|
| 110 |
from langchain_community.document_loaders import PDFMinerLoader
|
| 111 |
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
|
|
|
| 134 |
# Helper Functions
|
| 135 |
|
| 136 |
def extract_text_from_pdf(file_path):
|
| 137 |
+
"""Extract text from a PDF using PyMuPDF (fitz)."""
|
| 138 |
try:
|
| 139 |
+
doc = fitz.open(file_path)
|
| 140 |
+
text = ""
|
| 141 |
+
for page_num in range(doc.page_count):
|
| 142 |
+
page = doc.load_page(page_num)
|
| 143 |
+
text += page.get_text("text")
|
| 144 |
+
return text
|
| 145 |
except Exception as e:
|
| 146 |
logging.error(f"Error reading PDF {file_path}: {e}")
|
| 147 |
return None
|
|
|
|
| 160 |
file_path = os.path.join(uploaded_files_dir, filename)
|
| 161 |
logging.info(f"Processing file: {file_path}")
|
| 162 |
|
| 163 |
+
# Extract text using PyMuPDF
|
| 164 |
text = extract_text_from_pdf(file_path)
|
| 165 |
|
| 166 |
if text:
|