tensorplay.functional.take

tensorplay.functional.take(input, index) → Tensor[source]

Returns a new tensor with the elements of input at the given indices. The input tensor is treated as if it were viewed as a 1-D tensor. The result takes the same shape as the indices.

Parameters:

• input (Tensor) – the input tensor.

• index (LongTensor) – the indices into tensor

Example:

>>> src = tensorplay.tensor([[4, 3, 5],
...                     [6, 7, 8]])
>>> tensorplay.take(src, tensorplay.tensor([0, 2, 5]))
tensor([ 4,  5,  8])