model.eval()
    val_loss = 0
    val_accuracy = 0

    for batch in val_dataloader:

        batch_token_ids = batch[0].to(device)
        batch_attention_mask = batch[1].to(device)
        batch_labels = batch[2].to(device)

        with torch.no_grad():
            (loss, logits) = model(
                batch_token_ids,
                attention_mask = batch_attention_mask,
                labels = batch_labels,
                token_type_ids = None,
                return_dict=False)

        logits = logits.detach().cpu().numpy()
        label_ids = batch_labels.to('cpu').numpy()
        val_loss += loss.item()
        val_accuracy += calculate_accuracy(logits, label_ids)

    average_val_accuracy = val_accuracy / len(val_dataloader)