Skip to content

CosFaceLoss

CosFace: Large Margin Cosine Loss for Deep Face Recognition

losses.CosFaceLoss(num_classes, embedding_size, margin=0.35, scale=64,
    *, weight_regularizer=None, weight_reg_weight=1.0, weight_init_func=None, **options)

Equation:

\[ L_{lmc} = \frac{1}{N} \sum_{i} -\log \frac{e^{s \left(\cos\left(\theta_{y_i,i}\right) - m\right)}}{e^{s \left(\cos\left(\theta_{y_i,i}\right) - m\right)} + \sum_{j \neq y_i} e^{s \cos\left(\theta_{j,i}\right)}} \]

Here \(N\) is the batch size and \(\theta_{j,i}\) is the angle between embedding \(i\) and the weight vector of class \(j\).

Implementation notes: With an indices_tuple, each embedding's loss is multiplied by its miner weight. get_logits(embeddings) returns \(s \cos\theta_{j,i}\) without the margin; see classification logits. This loss derives from BaseProxyLoss.

Example:

import torch
from pytorch_metric_learning import losses

loss_func = losses.CosFaceLoss(num_classes=10, embedding_size=128)
optimizer = torch.optim.SGD(loss_func.parameters(), lr=0.01)
embeddings = torch.randn(32, 128, requires_grad=True)
labels = torch.randint(0, 10, (32,))
loss = loss_func(embeddings, labels)
loss.backward()
optimizer.step()

Parameters:

  • num_classes: The number of classes in the training set. Labels must be in 0..num_classes - 1.
  • embedding_size: The size of the embeddings passed to the loss.
  • margin: The cosine margin \(m\). The paper uses values between 0.25 and 0.45.
  • scale: The scale \(s\). The paper uses 64.
  • weight_regularizer, weight_reg_weight, weight_init_func: See learnable weight arguments. The class weights are weight, shaped (num_classes, embedding_size) and initialized with torch.nn.init.normal_. Include loss_func.parameters() in an optimizer.

Default distance:

  • CosineSimilarity(). This is the only compatible distance; another distance class raises TypeError.

Default reducer:

Reducer input:

  • loss: One loss per embedding, indexed by embedding position.