Update app.py

app.py

@@ -198,7 +198,7 @@ lm_datasets = tokenized_datasets.map(
 
 # Batches von Daten zusammenfassen
 tokenizer.pad_token = tokenizer.eos_token
-data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+#data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
 
 
 print ("###############lm datasets####################")
@@ -264,8 +264,8 @@ trainer = Trainer(
         args=training_args,
         train_dataset=lm_datasets["train"],
         eval_dataset=lm_datasets["test"],
-        data_collator=data_collator,
-        #tokenizer=tokenizer,
+        #data_collator=data_collator,
+        tokenizer=tokenizer,
         compute_metrics=compute_metrics,
 )