Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,18 +1,18 @@
-
 import streamlit as st
 import tensorflow as tf
 import numpy as np
 import re
 import nltk
-
-# Ensure NLTK sentence tokenizer is available
-nltk.download('punkt')
+import os
 from nltk.tokenize import sent_tokenize
 
-model = tf.keras.models.load_model('src/my_distilbert_classifier.keras')
-
-
+# Use local nltk_data folder (safe for Hugging Face Spaces)
+nltk_data_path = os.path.join(os.path.dirname(__file__), "nltk_data")
+nltk.download("punkt", download_dir=nltk_data_path)
+nltk.data.path.append(nltk_data_path)
 
+# Load model
+model = tf.keras.models.load_model('src/my_distilbert_classifier.keras')
 
 def predict_sentence_ai_probability(sentence):
     preds = model.predict([sentence])
@@ -20,7 +20,7 @@ def predict_sentence_ai_probability(sentence):
     return prob_ai
 
 def predict_ai_generated_percentage(text, threshold=0.75):
-    text=text+"."
+    text += "."
     sentences = sent_tokenize(text)
     ai_sentence_count = 0
     results = []
@@ -36,6 +36,7 @@ def predict_ai_generated_percentage(text, threshold=0.75):
     ai_percentage = (ai_sentence_count / total_sentences) * 100 if total_sentences > 0 else 0.0
     return ai_percentage, results
 
+# Streamlit UI
 st.title("🧠 AI Content Detector")
 st.markdown("This tool detects the percentage of **AI-generated content** in your input text based on sentence-level analysis.")