Update summarizer.py

updated with wordcloud and input option for both link and text

summarizer.py

@@ -1,7 +1,13 @@
 import re
+from extractdata import extract_text
+from wordcloud import plot_wordcloud
 from transformers import  AutoModelForSeq2SeqLM, AutoTokenizer
 
-def summarize(text, model):
+def summarize(input_, model):
+  if input_.split("/")[0] == "https:":
+    text = extract_text(input)
+  else:
+    text = input_
 
   if model == "T5":
     checkpoint = "csebuetnlp/mT5_multilingual_XLSum"
@@ -10,7 +16,6 @@ def summarize(text, model):
 
   WHITESPACE_HANDLER = lambda k: re.sub('\s+', ' ', re.sub('\n+', ' ', k.strip()))
 
-
   tokenizer = AutoTokenizer.from_pretrained(checkpoint)
   model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint)
 
@@ -33,6 +38,8 @@ def summarize(text, model):
   summary = tokenizer.decode(
                       output_ids,
                       skip_special_tokens=True,
-                      clean_up_tokenization_spaces=False )
-
-  return summary 
+                      clean_up_tokenization_spaces=False)
+                      
+  figure = plot_wordcloud(text)
+  
+  return summary, figure