add sumy option

requirements.txt

@@ -1,4 +1,5 @@
-nltk
-pypdf
+nltk==3.6.7
+pyenchant==3.2.2
+pypdf==3.7.0
 transformers
-torch
+wordninja==2.0.0

src/nltk_summarizer.py

@@ -159,6 +159,3 @@ def _correct_summary(self) -> None:
  def export(self, filename: str) -> None:
  with open(filename, mode="w", encoding="utf-8") as f:
  f.write(self.summary)
-
- def extract_keypoints(self) -> None:
- pass

src/sumy_summarizer.py

@@ -0,0 +1,30 @@
+from sumy.parsers.plaintext import PlaintextParser
+from sumy.nlp.tokenizers import Tokenizer
+from sumy.summarizers.lsa import LsaSummarizer as Summarizer
+from sumy.nlp.stemmers import Stemmer
+from sumy.utils import get_stop_words
+
+from converter import PDFToTextConverter
+
+
+class PDFSummarizer(PDFToTextConverter):
+
+ LANGUAGE = "english"
+ NUM_SENTENCES = 20
+
+ def __init__(self, filename) -> None:
+ super().__init__(filename)
+ self.summary = ""
+
+ def summarize(self) -> None:
+ stemmer = Stemmer(self.LANGUAGE)
+ summarizer = Summarizer(stemmer)
+ parser = PlaintextParser.from_string(self.text,
+ Tokenizer(self.LANGUAGE))
+ summarizer.stop_words = get_stop_words(self.LANGUAGE)
+ for sentence in summarizer(parser.document, self.NUM_SENTENCES):
+ self.summary += sentence._text
+
+ def export(self, filename: str) -> None:
+ with open(filename, mode="w", encoding="utf-8") as f:
+ f.write(self.summary)

src/test.py

@@ -1,7 +1,7 @@
 import argparse
 import nltk_summarizer
 import transformers_summarizer
-import converter
+import sumy_summarizer
 
 def main():
  parser = argparse.ArgumentParser(description='Get input .pdf')
@@ -28,11 +28,14 @@ def main():
  summarizer_NTLK = nltk_summarizer.PDFSummarizer(args.file)
  summarizer_NTLK.summarize()
  summarizer_NTLK.export(args.output)
- # elif args.mode == "pegasus":
- # summarizer_pegasus = transformers_summarizer.PDFSummarizer(args.file)
- # summarizer_pegasus.summarize()
- # summarizer_pegasus.export(args.output)
-
+ elif args.mode == "pegasus":
+ summarizer_pegasus = transformers_summarizer.PDFSummarizer(args.file)
+ summarizer_pegasus.summarize()
+ summarizer_pegasus.export(args.output)
+ elif args.mode == "sumy":
+ summarizer_sumy = sumy_summarizer.PDFSummarizer(args.file)
+ summarizer_sumy.summarize()
+ summarizer_sumy.export(args.output)
 
 if __name__ == "__main__":
  main()