Update app.py

app.py

@@ -1,5 +1,5 @@
 import gradio as gr
-from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
+from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline, AutoModelForSeq2SeqLM
 from pdfminer.high_level import extract_text
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.metrics.pairwise import cosine_similarity
@@ -74,17 +74,18 @@ def answer_question(pdf, question, num_words):
         elif relevant_text and len(answer.split()) < num_words:
             remaining_words = num_words - len(answer.split())
             added_words = 0
-            remaining_sentences = sent_tokenize(relevant_text)
-            for sentence in remaining_sentences:
+            added_sentences =
+            for sentence in sent_tokenize(relevant_text):
                 sentence_words = sentence.split()
                 words_to_add = min(remaining_words - added_words, len(sentence_words))
                 if words_to_add > 0:
-                    answer_words = answer.split()
-                    answer_words.extend(sentence_words[:words_to_add])
-                    answer = " ".join(answer_words)
+                    added_sentences.append(" ".join(sentence_words[:words_to_add]))
                     added_words += words_to_add
                 if added_words >= remaining_words:
                     break
+            answer += " " + " ".join(added_sentences)
+            answer = answer.strip()
+            answer = " ".join(answer.split())
             if len(answer.split()) > num_words:
                 answer = " ".join(answer.split()[:num_words])
 
@@ -96,10 +97,10 @@ def answer_question(pdf, question, num_words):
 with gr.Blocks() as iface:
     gr.Markdown("PDF Q&A with RoBERTa")
     with gr.Row():
-        with gr.Column(scale=2): # Question area takes up 2/3 of the row
+        with gr.Column(scale=2):
             question_input = gr.Textbox(lines=2, placeholder="Ask a question", label="Question")
             btn = gr.Button("Submit")
-        with gr.Column(scale=1): # Other inputs take the remaining 1/3
+        with gr.Column(scale=1):
             pdf_input = gr.File(type="filepath", label="Upload PDF")
             num_words_slider = gr.Slider(minimum=1, maximum=500, value=100, step=1, label="Number of Words")
     answer_output = gr.Textbox(label="Answer", lines=5)