Upload 2 files

app.py

@@ -36,6 +36,7 @@ logging.basicConfig(
 
 import gradio as gr
 import nltk
+nltk.download('punkt')  # 自動下載 tokenizer 資源
 import torch
 from cleantext import clean
 from doctr.models import ocr_predictor
@@ -663,4 +664,4 @@ if __name__ == "__main__":
             inputs=[summary_text, text_file],
             outputs=[aggregated_summary],
         )
-    demo.launch(share=args.share, debug=True)
+    demo.launch(share=args.share, debug=True)

requirements.txt

@@ -1,7 +1,9 @@
-transformers>=4.41.0
-sentencepiece
-torch
-pdf2image
-poppler-utils
-pytesseract
-Pillow
+torch>=2.0.0
+    torchvision>=0.15.0
+    transformers>=4.46.0
+    gradio>=4.24.0
+    pdf2image
+    pytesseract
+    nltk
+    Pillow
+    numpy