pervision (`bool`, *optional*):
            Whether to use the answer as the only supervision for aggregation examples.
        num_aggregation_labels (`int`, *optional*, defaults to 0):
            The number of aggregation operators to predict.

    Returns:
        aggregation_loss_known (`torch.FloatTensor` of shape `(batch_size,)`): Aggregation loss (when its type is known
        during training) per example.
    rD