Unit 4. build a music genre classifier

Hello, I followed all steps to fine tune the “ntu-spml/distilhubert” model for audio classification task and encountered this error:

trainer = Trainer(
model,
training_args,
train_dataset=gtzan_encoded[“train”],
eval_dataset=gtzan_encoded[“test”],
tokenizer=feature_extractor,
compute_metrics=compute_metrics,
)
trainer.train()

RuntimeError: “nll_loss_forward_reduce_cuda_kernel_2d_index” not implemented for ‘Int’

Did anybody get the same error? Please help with a solution. Thank you.