From the course: Hands-On Introduction to PyTorch for Machine Learning
Join today to access over 24,900 courses taught by industry experts.
Dataset exploration with PyTorch - PyTorch Tutorial
From the course: Hands-On Introduction to PyTorch for Machine Learning
Dataset exploration with PyTorch
- [Instructor] PyTorch provides a robust set of tools for data exploration, especially when preparing data sets for deep learning. These tools help users inspect, visualize, transform, and manage data sets efficiently before feeding them into models. Here's an overview of the most important ones. First of all, PyTorch provides two data primitives, torch.utils.data.dataset and torch.utils.data.dataloader. The dataset class stores the samples and the corresponding annotation labels. DataLoader efficiently loads data in batches with options for shuffling, parallel loading and more. These two classes are used for loading and inspecting batches of data, prototyping, and feeding data into models. PyTorch comes with a set of domain libraries that provides many preloaded datasets. There are subclasses of torch.utils.data.dataset. Examples include torchvision.datasets, torchtext.dataset, torchaudio.dataset. These preloaded sets and…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.