Update app.py

app.py

@@ -8,29 +8,29 @@ Original file is located at
 
 """  
 
-from sentence_splitter import SentenceSplitter, split_text_into_sentences
-from parrot import Parrot
-import warnings
-warnings.filterwarnings("ignore")
-
-#parrot = Parrot(model_tag="prithivida/parrot_paraphraser_on_T5")
+# importing the libraries
 import os
+import pandas as pd
 from parrot import Parrot
 import torch
 import warnings
 import nltk
-  
-#!pip install sentence-splitter   
-
+import warnings
 warnings.filterwarnings("ignore")
+
 splitter = SentenceSplitter(language='en')
+from sentence_splitter import SentenceSplitter, split_text_into_sentences
+from parrot import Parrot
 from transformers import PegasusForConditionalGeneration, PegasusTokenizer
 from transformers import AutoTokenizer
 from transformers import AutoModelForSeq2SeqLM
-import pandas as pd
+
 from parrot.filters import Adequacy
 from parrot.filters import Fluency
 from parrot.filters import Diversity
+
+# Adding the metrics
+
 adequacy_score = Adequacy()
 fluency_score  = Fluency()
 diversity_score= Diversity()
@@ -38,7 +38,11 @@ device= "cuda:0"
 adequacy_threshold = 0.90
 fluency_threshold = 0.90
 diversity_ranker="levenshtein"
-do_diverse=True
+do_diverse=False
+#num_beam_groups=num_beams, diversity_penalty=0.5
+
+# adding the model
+
 model_name = 'tuner007/pegasus_paraphrase'
 torch_device = 'cuda' if torch.cuda.is_available() else 'cpu'
 tokenizer = PegasusTokenizer.from_pretrained(model_name)
@@ -48,7 +52,7 @@ def get_max_str(lst):
     return max(lst, key=len)
 def get_response(input_text,num_return_sequences=10,num_beams=10):
     batch = tokenizer.prepare_seq2seq_batch([input_text],truncation=True,padding='longest',max_length=30,return_tensors='pt').to(torch_device)
-    translated = model_pegasus.generate(**batch,max_length=30,num_beams=num_beams, num_return_sequences=num_return_sequences, num_beam_groups=num_beams, diversity_penalty=0.5, temperature=1.5)
+    translated = model_pegasus.generate(**batch,max_length=30,num_beams=num_beams, num_return_sequences=num_return_sequences, temperature=1.5)
     tgt_text = tokenizer.batch_decode(translated, skip_special_tokens=True)
     try:
         adequacy_filtered_phrases = adequacy_score.filter(input_text,tgt_text, adequacy_threshold, device)
@@ -63,6 +67,8 @@ def get_response(input_text,num_return_sequences=10,num_beams=10):
             return get_max_str(adequacy_filtered_phrases)
     except:
         return(get_max_str(tgt_text))
+        
+# Deploying the model
 
 import gradio as gr