January 2019
Intermediate to advanced
386 pages
11h 13m
English
book
Python Deep Learning - Second Edition
by Ivan Vasilev, Daniel Slater, Gianmario Spacagna, Peter Roelants, Valentino Zocca
Content preview from Python Deep Learning - Second Edition
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
Start your free trial
Transfer learning example with PyTorch
Now that we know what transfer learning is, let's see whether it works in practice. In this section, we'll apply an advanced ImageNet pre-trained network on the CIFAR-10 images. We'll use both types of transfer learning. It's preferable to run this example on GPU:
- Do the following imports:
import torchimport torch.nn as nnimport torch.optim as optimimport torchvisionfrom torchvision import models, transforms
- Define batch_size for convenience:
batch_size = 50
- Define the training dataset. We have to consider a few things:
- The CIFAR-10 images are 32 x 32, while the ImageNet network expects 224 x 224 input. As we are using ImageNet based network, we'll upsample the 32x32 CIFAR images to 224x224.