If you use Pytorch (https://pytorch.org) you can use the implemented models in torchvision (https://pytorch.org/docs/stable/torchvision/models.html):
import torchvision.models as models alexnet = models.alexnet()
Also, you use the parameter "pretrained =True" for a model pre-trained on ImageNet
For the data set you can use the implemented for Pytorch in https://pytorch.org/docs/stable/torchvision/datasets.html?highlight=dataset%20image#torchvision.datasets.ImageFolder.