Update app.py

app.py

@@ -33,14 +33,14 @@ def load_and_prepare_dataset():
         return tokenizer(examples["text"], truncation=True, max_length=512)
     
     tokenized_datasets = subset.map(tokenize_function, batched=True)
-    return tokenized_datasets.train_test_split(test_size=0.1, seed=42)
+    return tokenized_datasets
 
-split_dataset = load_and_prepare_dataset()
+tokenized_dataset = load_and_prepare_dataset()
 
 # === 4️⃣ EĞİTİM AYARLARI ===
 batch_size = 1  
 num_epochs = 1  
-max_steps = (len(split_dataset["train"]) // batch_size) * num_epochs
+max_steps = (len(tokenized_dataset) // batch_size) * num_epochs
 
 training_args = TrainingArguments(
     output_dir="./mistral_lora",
@@ -60,8 +60,7 @@ training_args = TrainingArguments(
 trainer = Trainer(
     model=model,
     args=training_args,
-    train_dataset=split_dataset["train"],
-    eval_dataset=split_dataset["test"],
+    train_dataset=tokenized_dataset,
 )
 
 @spaces.GPU
@@ -86,4 +85,4 @@ demo = gr.ChatInterface(
 
 if __name__ == "__main__":
     train_model()
-    demo.launch(share=True)
+    demo.launch(share=True)