pad2d8

torch_brain.batching.pad2d8(obj)

Wrap an object to specify that it (or any of its members) should be padded to the maximum length in the batch along its first two dimensions. This function is similar to pad2d except that the inner padding length is rounded up to the nearest multiple of 8.

Parameters:

obj – Can be tensors, numpy arrays, lists, tuples, or dictionaries.