Trait TopKLastDimOp 

pub trait TopKLastDimOp {
    // Required methods
    fn topk(&self, topk: usize) -> Result<TopKOutput>;
    fn topk_unsorted(&self, topk: usize) -> Result<TopKOutput>;
}

Required Methods

fn topk(&self, topk: usize) -> Result<TopKOutput>

Topk in the last dim. values retains a gradient but indices has none w.r.t self. This expects a contiguous tensor. Note: this implements torch.topk with sorted=True.

fn topk_unsorted(&self, topk: usize) -> Result<TopKOutput>

Topk in the last dim. values retains a gradient but indices has none w.r.t self. This expects a contiguous tensor. Note: this implements torch.topk with sorted=False.

Implementations on Foreign Types

impl TopKLastDimOp for Tensor

fn topk(&self, topk: usize) -> Result<TopKOutput>

fn topk_unsorted(&self, topk: usize) -> Result<TopKOutput>

Implementors