ResNet network architecture. Designed for complex datasets.

expand + depthwise + pointwise

expansion + depthwise + pointwise + squeeze-excitation

expand + depthwise + pointwise

expansion + depthwise + pointwise + squeeze-excitation

Market1501 Reference: Zheng et al. Scalable Person Re-identification: A Benchmark. ICCV 2015. URL: http://www.liangzheng.org/Project/

Squeeze-and-Excitation block with Swish.

Squeeze-and-Excitation block with Swish.

Base class of reid dataset

Base class of image reid dataset

The basic block of ResNet.

Federated learning Dataset setting.

Federated learning model.

Supervised Contrastive Learning: https://arxiv.org/pdf/2004.11362.pdf. It also supports the unsupervised contrastive loss in SimCLR