Upload preprocess.py

preprocess.py

@@ -20,6 +20,16 @@ try:
 except Exception as e:
     logger.warning(f"NLTK data download failed: {e}")
 
+# Guarded NLTK downloads
+if hasattr(nltk, "download"):
+    try:
+        nltk.download('punkt', quiet=True)
+        nltk.download('averaged_perceptron_tagger', quiet=True)
+    except Exception as e:
+        logger.warning(f"NLTK download failed: {e}")
+else:
+    logger.warning("NLTK.download not available; skipping corpus downloads")
+
 def get_tokenizer_wrapper():
     try:
         tokenizer = get_tokenizer("bert-base-uncased")