Update app.py

app.py

@@ -1,24 +1,18 @@
 import streamlit as st
-import nltk
 import pandas as pd
 from py_thesaurus import Thesaurus
 import random
+from transformers import GPT2LMHeadModel, GPT2Tokenizer
 
-# Cache the NLTK corpora and functions for better performance
-@st.cache(allow_output_mutation=True)
-def load_nltk_data():
-    nltk.download('gutenberg')
-    from nltk.corpus import gutenberg
+# Load GPT-2 model and tokenizer
+model = GPT2LMHeadModel.from_pretrained("gpt2")
+tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
 
-    return gutenberg.words('austen-emma.txt')
-
-# Function to generate random text from NLTK corpora
-def generate_text():
-    text = load_nltk_data()
-    words = [word.lower() for word in text if word.isalpha()]
-    random_text = ' '.join(nltk.choice(words) for _ in range(10))
-
-    return random_text
+# Function to generate random text using GPT-2
+def generate_text(prompt):
+    inputs = tokenizer.encode(prompt, return_tensors="pt")
+    outputs = model.generate(inputs, max_length=50, num_return_sequences=1)
+    return tokenizer.decode(outputs[0])
 
 # Function to replace a random word with its synonym
 def replace_with_synonym(sentence):
@@ -60,8 +54,9 @@ st.write(scoreboard)
 # Save scoreboard to output CSV file
 scoreboard.to_csv('output.csv', index=False)
 
-# Generate a random text and replace a word with a synonym
-original_text = generate_text()
+# Generate a random text using GPT-2
+prompt = "Tell me a joke:"
+original_text = generate_text(prompt)
 modified_text = replace_with_synonym(original_text)
 
 # Display the modified text