Update app.py

app.py

@@ -14,9 +14,9 @@ tokenizer = GPT2Tokenizer.from_pretrained(model_name)
 model = GPT2LMHeadModel.from_pretrained(model_name)
 
 # Cargar datasets y mantener todo en RAM
-dataset_humanizado = load_dataset('daily_dialog', split='train')
-dataset_codigo = load_dataset('code_search_net', split='train')
-dataset_prompts = load_dataset('openai_humaneval', split='train')
+dataset_humanizado = load_dataset('daily_dialog', split='train', trust_remote_code=True)
+dataset_codigo = load_dataset('code_search_net', split='train', trust_remote_code=True)
+dataset_prompts = load_dataset('openai_humaneval', split='train', trust_remote_code=True)
 
 combined_dataset = concatenate_datasets([
     dataset_humanizado,
@@ -31,7 +31,7 @@ def tokenize_function(examples):
 tokenized_dataset = combined_dataset.map(tokenize_function, batched=True)
 
 training_args = TrainingArguments(
-    output_dir='./results',  # Puede ser usado para guardar resultados, pero no es necesario en RAM
+    output_dir='./results',
     per_device_train_batch_size=100,
     per_device_eval_batch_size=100,
     num_train_epochs=0,
@@ -68,4 +68,4 @@ def run_training():
 run_training()
 
 import shutil
-shutil.rmtree('./results', ignore_errors=True)  # Limpiar si es necesario, aunque puede no ser requerido si todo está en RAM
+shutil.rmtree('./results', ignore_errors=True)