PFLlib

Models

We currently support multiple models for various datasets, some of which include pre-trained model weights. Please refer to the code for more details.

• for MNIST and Fashion-MNIST
  • Mclr_Logistic(1*28*28) # convex
  • LeNet()
  • DNN(1*28*28, 100)
• for Cifar10, Cifar100 and Tiny-ImageNet
  • Mclr_Logistic(3*32*32) # convex
  • FedAvgCNN()
  • DNN(3*32*32, 100)
  • ResNet18, AlexNet, MobileNet, GoogleNet, etc.
• for AG_News and Sogou_News
  • LSTM()
  • fastText() in Bag of Tricks for Efficient Text Classification
  • TextCNN() in Convolutional Neural Networks for Sentence Classification
  • TransformerModel() in Attention is all you need
• for AmazonReview
  • AmazonMLP() in Curriculum manager for source selection in multi-source domain adaptation
• for Omniglot
  • FedAvgCNN()
• for HAR and PAMAP
  • HARCNN() in Convolutional neural networks for human activity recognition using mobile sensors