Functions
	concat_all_gather (tensor)

	normalize (x, axis=-1)

	euclidean_dist (x, y)

	cosine_dist (x, y)

Detailed Description

@author:  xingyu liao
@contact: sherlockliao01@gmail.com

Function Documentation

◆ concat_all_gather()

fastreid.modeling.losses.utils.concat_all_gather ( tensor )

Performs all_gather operation on the provided tensors.
*** Warning ***: torch.distributed.all_gather has no gradient.

Definition at line 10 of file utils.py.

def concat_all_gather(tensor):
    """
    Performs all_gather operation on the provided tensors.
    *** Warning ***: torch.distributed.all_gather has no gradient.
    """
    tensors_gather = [torch.ones_like(tensor)
                      for _ in range(torch.distributed.get_world_size())]
    torch.distributed.all_gather(tensors_gather, tensor, async_op=False)
 
    output = torch.cat(tensors_gather, dim=0)
    return output
 
 

◆ cosine_dist()

fastreid.modeling.losses.utils.cosine_dist	(		x,
			y )

Definition at line 43 of file utils.py.

def cosine_dist(x, y):
    bs1, bs2 = x.size(0), y.size(0)
    frac_up = torch.matmul(x, y.transpose(0, 1))
    frac_down = (torch.sqrt(torch.sum(torch.pow(x, 2), 1))).view(bs1, 1).repeat(1, bs2) * \
                (torch.sqrt(torch.sum(torch.pow(y, 2), 1))).view(1, bs2).repeat(bs1, 1)
    cosine = frac_up / frac_down
    return 1 - cosine

◆ euclidean_dist()

fastreid.modeling.losses.utils.euclidean_dist	(		x,
			y )

Definition at line 34 of file utils.py.

def euclidean_dist(x, y):
    m, n = x.size(0), y.size(0)
    xx = torch.pow(x, 2).sum(1, keepdim=True).expand(m, n)
    yy = torch.pow(y, 2).sum(1, keepdim=True).expand(n, m).t()
    dist = xx + yy - 2 * torch.matmul(x, y.t())
    dist = dist.clamp(min=1e-12).sqrt()  # for numerical stability
    return dist
 
 

◆ normalize()

fastreid.modeling.losses.utils.normalize	(		x,
			axis = -1 )

Normalizing to unit length along the specified dimension.
Args:
  x: pytorch Variable
Returns:
  x: pytorch Variable, same shape as input

Definition at line 23 of file utils.py.