torch.nn.utils.rnn.unpad_sequence

torch.nn.utils.rnn.unpad_sequence(padded_sequences, lengths, batch_first=False)

将填充的 Tensor 填充成不同长度的 Tensor 列表。

将填充的 Tensor 拆分为一个变长 Tensor 列表。

示例

>>> from torch.nn.utils.rnn import pad_sequence, unpad_sequence
>>> a = torch.ones(25, 300)
>>> b = torch.ones(22, 300)
>>> c = torch.ones(15, 300)
>>> sequences = [a, b, c]
>>> padded_sequences = pad_sequence(sequences)
>>> lengths = torch.as_tensor([v.size(0) for v in sequences])
>>> unpadded_sequences = unpad_sequence(padded_sequences, lengths)
>>> torch.allclose(sequences[0], unpadded_sequences[0])
True
>>> torch.allclose(sequences[1], unpadded_sequences[1])
True
>>> torch.allclose(sequences[2], unpadded_sequences[2])
True

参数:

• padded_sequences (Tensor) – 填充序列。

• lengths (Tensor) – 原始（未填充）序列的长度。

• batch_first (bool, 可选) – 是否批处理维度在前。默认值： False 。

返回值:

Tensor 对象列表

返回类型:

list[torch.Tensor]