Computes and stores the average and current value

Computes and stores the average and current value

Computes and stores the average and current value

Implement of Am_softmax (https://arxiv.org/pdf/1801.05599.pdf): Args: in_features: size of each input sample out_features: size of

r"""Implement of ArcFace (https://arxiv.org/pdf/1801.07698v1.pdf): Args: in_features: size of each input sample out

Implement of ArcFace (https://arxiv.org/pdf/1801.07698v1.pdf): Args: in_features: size of each input sample out_featur

A named tuple describing a ResNet block.

A named tuple describing a ResNet block.

Implement of CosFace (https://arxiv.org/pdf/1801.09414.pdf): Args: in_features: size of each input sample out_features: size of ea

Mobile Inverted Residual Bottleneck Block. Args: block_args (namedtuple): BlockArgs, defined in utils.py. global_params (namedtup

Fix the backbones variable names for pretrained weight

A simple structure that contains basic shape specification about a tensor. It is often used as the auxiliary inputs/outputs of models, to

Implement of Softmax (normal classification head): Args: in_features: size of each input sample out_features: size of

Implement of SphereFace (https://arxiv.org/pdf/1704.08063.pdf): Args: in_features: size of each input sample out_features: size of

r"""Implementation for "Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations" Args: in_features: size of

r"""Implement of Am_softmax (https://arxiv.org/pdf/1801.05599.pdf): Args: in_features: size of each input sample out_features:

r"""Implement of Am_softmax (https://arxiv.org/pdf/1704.06369.pdf): Args: in_features: size of each input sample out_features:

r"""Implement of ArcFace (https://arxiv.org/pdf/1801.07698v1.pdf): Args: in_features: size of each input sample out

Implement of Towards Flops-constrained Face Recognition (https://arxiv.org/pdf/1909.00632.pdf):

r"""Implement of CosFace (https://arxiv.org/pdf/1801.09414.pdf): Args: in_features: size of each input sample out_features: siz

r"""Implement of CosFace (https://arxiv.org/pdf/1801.09414.pdf): Args: in_features: size of each input sample out_features: siz

Implementation for "CurricularFace: Adaptive Curriculum Learning Loss for Deep Face Recognition".

Implementation for "ArcFace: Additive Angular Margin Loss for Deep Face Recognition"

r"""Implement of SphereFace (https://arxiv.org/pdf/1704.08063.pdf): Args: in_features: size of each input sample out_features:

r"""Implement of SphereFace (https://arxiv.org/pdf/1704.08063.pdf): Args: in_features: size of each input sample out_features:

Implementation for "Additive Margin Softmax for Face Verification"

Block Decoder for readability, straight from the official TensorFlow repository.

Implementation for "Circle Loss: A Unified Perspective of Pair Similarity Optimization" Note: this is the classification based implementation of

2D Convolutions like TensorFlow, for a dynamic image size. The padding is operated in forward function by calculating dynamically.

2D Convolutions like TensorFlow's 'SAME' mode, with the given input image size. The padding mudule is calculated in construction function, then

EfficientNet model. Most easily loaded with the .from_name or .from_pretrained methods. Args: blocks_args (list[namedtuple]): A li

Ghost bottleneck w/ optional SE

Implementation for "Mis-classified Vector Guided Softmax Loss for Face Recognition"

2D MaxPooling like TensorFlow's 'SAME' mode, with a dynamic image size. The padding is operated in forward function by calculating dynamically.

2D MaxPooling like TensorFlow's 'SAME' mode, with the given input image size. The padding mudule is calculated in construction function, then u

Implementation for "NPCFace: A Negative-Positive Cooperation Supervision for Training Large-scale Face Recognition"

Implementation for "Semi-Siamese Training for Shallow Face Learning".

r"""Implement of Softmax (normal classification head): Args: in_features: size of each input sample out_features: s

r"""Implement of Softmax (normal classification head): Args: in_features: size of each input sample out_features: s