FedAWAC

CIFAR-10 and CIFAR-100 datasets comprise images of $10$ and $100$ categories, respectively, with a fixed size of 32x32 pixels in color. We train a Convolutional Neural Network (CNN) model to classify images, which consists of $2$ convolutional layers (5x5, each activated by ReLU and followed by 2\x2 max pooling), 2 fully connected layers, and Softmax normalizes the final output.

CIFAR-10与CIFAR-100数据集分别包含10个和100个类别的图像，所有图像均为固定尺寸32×32像素的彩色图像。我们将训练一个卷积神经网络（Convolutional Neural Network，CNN）模型以完成图像分类任务，该模型包含2个卷积层（卷积核尺寸为5×5，每一层均通过ReLU激活函数激活，并随后接入2×2最大池化操作）、2个全连接层，最终输出将由Softmax函数完成归一化处理。