Update app.py

app.py

@@ -8,28 +8,6 @@
 # PDF
 # -------------------------
 
-#!pip install PyPDF2
-#!pip install pdfminer.six
-#!pip install pdfplumber
-#!pip install pdf2image
-#!pip install Pillow
-#!pip install pytesseract
-#!pip install poppler-utils
-#!pip install tesseract-ocr
-#!pip install libtesseract-dev
-
-#!pip install fastapi
-#!pip install -q torch
-#!pip install -q transformers
-#!pip install -q gradio
-#!pip install ffmpeg
-
-
-#!apt-get install poppler-utils
-#!apt install tesseract-ocr
-#!apt install libtesseract-dev
-
-
 # To read the PDF
 import PyPDF2
 # To analyze the PDF layout and extract text
@@ -281,35 +259,6 @@ pdf_path=os.path.join(os.path.abspath(""), "hidden-technical-debt-in-machine-lea
 pdf_path2=os.path.join(os.path.abspath(""), "1812_05944.pdf")
 
 
-text_per_page = read_pdf(pdf_path)
-
-text_per_page.keys()
-
-
-page_1 = text_per_page['Page_0']
-
-# ============================================================================================
-
-# picking up the abstract from the first page content
-flag=False
-abstract_sect=""
-
-for i in range(len(page_1)):
-  if page_1[0][i].strip()=="Abstract":
-    flag=True
-  if page_1[0][i].strip()=="1 Introduction":
-    flag = False
-  if flag:
-    # abstract_sect contains the Abstract section content
-    abstract_sect+=page_1[0][i]
-
-
-from transformers import pipeline
-
-summarizer = pipeline("summarization", model="knkarthick/MEETING_SUMMARY")
-summary=(summarizer(abstract_sect))
-summary_text=summary[0].get("summary_text")
-print(summary_text)
 
 
 
@@ -333,8 +282,39 @@ def sentence_to_audio(summary_txt):
     return sampling_rate, speech_values.cpu().numpy().squeeze()
 
 
-#summary_txt="It is dangerous to think of machine learning as a free-to-use toolkit, as it is common to incur ongoing maintenance costs in real-world ML systems"
-sentence_to_audio(summary_text)
+text_per_page = read_pdf(pdf_path)
+text_per_page.keys()
+page_1 = text_per_page['Page_0']
+
+# ============================================================================================
+
+# picking up the abstract from the first page content
+#flag=False
+#abstract_sect=""
+
+#for i in range(len(page_1)):
+#  if page_1[0][i].strip()=="Abstract":
+#    flag=True
+#  if page_1[0][i].strip()=="1 Introduction":
+#    flag = False
+#  if flag:
+#    # abstract_sect contains the Abstract section content
+#    abstract_sect+=page_1[0][i]
+
+
+#from transformers import pipeline
+#
+#summarizer = pipeline("summarization", model="knkarthick/MEETING_SUMMARY")
+#summary=(summarizer(abstract_sect))
+#summary_text=summary[0].get("summary_text")
+#print(summary_text)
+
+
+# ===========================================================    
+
+summary_txt="It is dangerous to think of machine learning as a free-to-use toolkit, as it is common to incur ongoing maintenance costs in real-world ML systems"
+
+sentence_to_audio(summary_txt)
 
 pdf_path=os.path.join(os.path.abspath(""), "hidden-technical-debt-in-machine-learning-systems-Paper.pdf")
 pdf_path2=os.path.join(os.path.abspath(""), "1812_05944.pdf")