no eval

app.py

@@ -26,6 +26,7 @@ tokenized_dataset = dataset.map(preprocess_data, batched=True)
 training_args = TrainingArguments(
     output_dir="./results",
     evaluation_strategy="epoch",
+    eval_strategy="no",  # Disabilita la valutazione durante l'addestramento
     learning_rate=2e-5,
     per_device_train_batch_size=4,
     num_train_epochs=3,
@@ -39,13 +40,15 @@ trainer = Trainer(
     train_dataset=tokenized_dataset["train"]
 )
 
+trainer.train()
+
 # Funzione di Gradio per effettuare previsioni
-def answer_question(question):
-    inputs = tokenizer(question, return_tensors="pt", padding="max_length", truncation=True, max_length=128)
-    outputs = model.generate(inputs["input_ids"], max_length=50)
-    answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    return answer
-
-# Interfaccia Gradio
-iface = gr.Interface(fn=answer_question, inputs="text", outputs="text")
-iface.launch()
+#def answer_question(question):
+#    inputs = tokenizer(question, return_tensors="pt", padding="max_length", truncation=True, max_length=128)
+#    outputs = model.generate(inputs["input_ids"], max_length=50)
+#    answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
+#    return answer
+#
+## Interfaccia Gradio
+#iface = gr.Interface(fn=answer_question, inputs="text", outputs="text")
+#iface.launch()