Update app.py

app.py

@@ -1,9 +1,7 @@
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
 import torch
-import nltk
-
-nltk.download("punkt")
+import re
 
 # Load model
 MODEL = "roberta-base-openai-detector"
@@ -11,10 +9,13 @@ tokenizer = AutoTokenizer.from_pretrained(MODEL)
 model = AutoModelForSequenceClassification.from_pretrained(MODEL)
 
 def detect_ai(text):
-    sentences = nltk.sent_tokenize(text)
+    # Split into rough sentences
+    sentences = re.split(r'(?<=[.!?]) +', text)
     results = []
 
     for sent in sentences:
+        if not sent.strip():
+            continue
         inputs = tokenizer(sent, return_tensors="pt", truncation=True, max_length=512)
         with torch.no_grad():
             outputs = model(**inputs)