Update app.py

app.py

@@ -5,12 +5,12 @@ from datasets import load_dataset
 # 1️⃣ Modell & Tokenizer laden
 model_name = "allenai/scibert_scivocab_uncased"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=3)  # z.B. für 3 Kategorien
+model = AutoModelForSequenceClassification.from_pretrained(model_name, num_labels=3)
 
-# 2️⃣ Dataset laden (armanc/scientific_papers)
-dataset = load_dataset("armanc/scientific_papers")
+# 2️⃣ Dataset laden (armanc/scientific_papers) mit trust_remote_code=True
+dataset = load_dataset("armanc/scientific_papers", trust_remote_code=True)
 
-# 3️⃣ Tokenisierung der Texte
+# 3️⃣ Tokenisierung der Texte (hier wird die Spalte "text" genutzt; ggf. anpassen, falls andere Spalten vorhanden sind)
 def tokenize_function(examples):
     return tokenizer(examples["text"], padding="max_length", truncation=True)