niteshade.models

Abstract base model class for niteshade workflows as well as some specific toy models for out-of-the-box use.

Classes

BaseModel(architecture, optimizer, loss_func, lr)	Abstract model class intended for ease of implementation in designing neural networks for data poisoning attacks.
CifarClassifier([optimizer, loss_func, lr, ...])	ResNet-18 classifier inheriting from BaseModel for the torchvision CIFAR10 dataset.
IrisClassifier([optimizer, loss_func, lr, ...])	Pre-defined simple classifier for the Iris dataset containing three fully-connected layers with neurons 4--50--3 using ReLU as an activation function and a Softmax output activation.
MNISTClassifier([optimizer, loss_func, lr, ...])	Pre-defined classifier for the MNIST dataset.