Practical: Custom datasets
A link to a Colab article demonstrating how to make custom datasets in PyTorch.
© The University of Nottingham
To use new image data we will need to make a custom dataset.
In the linked Colab notebook we will make a custom dataset in PyTorch based on the Oxford flower dataset.
In the notebook we will cover the following:
- downloading the dataset
- making a custom dataset class
- splitting into training and validation data
- image transformations – data augmentation.
Follow the link below and work through the Colab notebook step by step.
You may wish to open the link in a new browser tab so you can refer back here quickly.
Please leave any questions or comments below.
Reach your personal and professional goals
Unlock access to hundreds of expert online courses and degrees from top universities and educators to gain accredited qualifications and professional CV-building certificates.
Join over 18 million learners to launch, switch or build upon your career, all at your own pace, across a wide range of topic areas.
