Update app.py

app.py

@@ -1,4 +1,39 @@
+# import torch
+# from transformers import PegasusForConditionalGeneration, PegasusTokenizer
 
+# model_name = 'tuner007/pegasus_paraphrase'
+# torch_device = 'cuda' if torch.cuda.is_available() else 'cpu'
+# tokenizer = PegasusTokenizer.from_pretrained(model_name)
+# model = PegasusForConditionalGeneration.from_pretrained(model_name).to(torch_device)
+
+# def get_response(input_text,num_return_sequences):
+#   batch = tokenizer.prepare_seq2seq_batch([input_text],truncation=True,padding='longest',max_length=60, return_tensors="pt").to(torch_device)
+#   translated = model.generate(**batch,max_length=60,num_beams=10, num_return_sequences=num_return_sequences, temperature=1.5)
+#   tgt_text = tokenizer.batch_decode(translated, skip_special_tokens=True)
+#   return tgt_text
+
+# from sentence_splitter import SentenceSplitter, split_text_into_sentences
+
+# splitter = SentenceSplitter(language='en')
+
+# def paraphraze(text):
+#   sentence_list = splitter.split(text)
+#   paraphrase = []
+
+#   for i in sentence_list:
+#     a = get_response(i,1)
+#     paraphrase.append(a)
+#     paraphrase2 = [' '.join(x) for x in paraphrase]
+#     paraphrase3 = [' '.join(x for x in paraphrase2) ]
+#   paraphrased_text = str(paraphrase3).strip('[]').strip("'")
+#   return paraphrased_text
+
+
+# def summarize(text):
+
+#   paraphrased_text = paraphraze(text)
+#   return paraphrased_text
+########################################################################################################
 
 import torch
 from transformers import PegasusForConditionalGeneration, PegasusTokenizer
@@ -15,39 +50,6 @@ hide_main_and_footer_style = """
 
 st.markdown(hide_main_and_footer_style, unsafe_allow_html=True)
 
-model_name = 'tuner007/pegasus_paraphrase'
-torch_device = 'cuda' if torch.cuda.is_available() else 'cpu'
-tokenizer = PegasusTokenizer.from_pretrained(model_name)
-model = PegasusForConditionalGeneration.from_pretrained(model_name).to(torch_device)
-
-def get_response(input_text,num_return_sequences):
-  batch = tokenizer.prepare_seq2seq_batch([input_text],truncation=True,padding='longest',max_length=60, return_tensors="pt").to(torch_device)
-  translated = model.generate(**batch,max_length=60,num_beams=10, num_return_sequences=num_return_sequences, temperature=1.5)
-  tgt_text = tokenizer.batch_decode(translated, skip_special_tokens=True)
-  return tgt_text
-
-from sentence_splitter import SentenceSplitter, split_text_into_sentences
-
-splitter = SentenceSplitter(language='en')
-
-def paraphraze(text, how_many=1):
-  sentence_list = splitter.split(text)
-  paraphrase = []
-
-  for i in sentence_list:
-    a = get_response(i,how_many)
-    paraphrase.append(a)
-    paraphrase2 = [' '.join(x) for x in paraphrase]
-    paraphrase3 = [' '.join(x for x in paraphrase2) ]
-  paraphrased_text = paraphrase #str(paraphrase3).strip('[]').strip("'")
-  return paraphrased_text
-
-
-def summarize(text):
-
-  paraphrased_text = paraphraze(text)
-  return paraphrased_text
-########################################################################################################
 
 from transformers import *
 
@@ -57,7 +59,7 @@ tokenizer = PegasusTokenizerFast.from_pretrained("tuner007/pegasus_paraphrase")
 
 def get_paraphrased_sentences(model, tokenizer, sentence, num_return_sequences=5, num_beams=5):
   # tokenize the text to be form of a list of token IDs
-  inputs = tokenizer([sentence], truncation=False, padding="longest", return_tensors="pt")
+  inputs = tokenizer([sentence], truncation=True, padding="longest", return_tensors="pt")
   # generate the paraphrased sentences
   outputs = model.generate(
     **inputs,
@@ -68,6 +70,8 @@ def get_paraphrased_sentences(model, tokenizer, sentence, num_return_sequences=5
   return tokenizer.batch_decode(outputs, skip_special_tokens=True)
 
 
+import streamlit as st
+#from .paraphraser import get_paraphrased_sentences, model, tokenizer
 
 
 def app():
@@ -78,24 +82,18 @@ def app():
     x = 0
     output = ['Result ']
     with col1:
-        user_input = st.text_area('Enter text','', height=200)
+        user_input = st.text_area('Enter text','', height=300)
         
         paraphraseNo = st.slider('Number of Parapharases',1,10,2)
-        if st.button('Single-Paraphrase'):
-            with st.spinner(text="This may take a moment..."):
-                output = summarize(user_input) 
-                
-        if st.button('Multiple-Paraphrase'):
+        if st.button('Paraphrase'):
             with st.spinner(text="This may take a moment..."):
                 output = get_paraphrased_sentences(model, tokenizer, user_input, num_beams=10, num_return_sequences=paraphraseNo)    
-                
-
-                
+    
     #with spacer:
     
     with col2:
         for x, element in enumerate(output):
-            user_output = st.text_area(label="", value=output[x], height=150 )
+            user_output = st.text_area(label="", value=output[x], height=200 )
 
 # st.markdown(
 #     '''<style>
@@ -104,7 +102,6 @@ def app():
 #     </style>''', unsafe_allow_html=True
 # )
 
-
 if __name__ == '__main__':
 	app()