track_mask8# torch_brain.batching.track_mask8(input)[source]# Wrap an array or tensor to specify that its padding mask should be tracked. This is used in conjunction with pad8. Parameters: input (Tensor | ndarray) – An array or tensor.
