fl_sim.models

This module contains built-in simple models.

Convolutional neural networks (CNN)

CNNMnist(num_classes)	Convolutional neural network using MNIST type input.
CNNFEMnist([num_classes])	Convolutional neural network using FEMnist type input.
CNNFEMnist_Tiny([num_classes])	Tiny version of CNNFEMnist.
CNNCifar(num_classes)	Convolutional neural network using CIFAR type input.
CNNCifar_Small(num_classes)	Convolutional neural network using CIFAR type input.
CNNCifar_Tiny(num_classes)	Convolutional neural network using CIFAR type input.
ResNet18(num_classes[, pretrained])	ResNet18 model for image classification.
ResNet10(num_classes[, pretrained])	ResNet10 model for image classification.

Recurrent neural networks (RNN)

RNN_OriginalFedAvg([embedding_dim, ...])	Creates a RNN model using LSTM layers for Shakespeare language models (next character prediction task).
RNN_StackOverFlow([vocab_size, ...])	Creates a RNN model using LSTM layers for StackOverFlow (next word prediction task).
RNN_Sent140([latent_size, num_classes, ...])	Stacked LSTM model for sentiment analysis on the Sent140 dataset.
RNN_Sent140_LITE([latent_size, num_classes, ...])	Stacked LSTM model for sentiment analysis on the Sent140 dataset.

Multilayer perceptron (MLP)

MLP(dim_in, dim_out[, dim_hidden, ...])	Multi-layer perceptron.
FedPDMLP(dim_in, dim_hidden, dim_out[, ndim])	Multi-layer perceptron modified from FedPD/models.py.

Linear models

LogisticRegression(num_features, num_classes)	Logistic regression model for classification task.
SVC(num_features, num_classes)	Support vector machine classifier.
SVR(num_features)	Support vector machine regressor.

Utilities

reset_parameters(module)	Reset the parameters of a module and its children.
top_n_accuracy(preds, labels[, n])	Top-n accuracy.
CLFMixin()	Mixin class for classifiers.
REGMixin()	Mixin for regressors.
DiffMixin()	Mixin for differences of two models.