Explore the essential components of data handling in PyTorch for deep learning tasks in this 30-minute video tutorial. Dive into the Fashion-MNIST dataset, master the usage of PyTorch's Dataset and DataLoader classes, and learn how to efficiently move tensors to CUDA for GPU-accelerated training. Discover techniques for transforming input data using PyTorch's transform functions and implement target transformations with lambda functions for one-hot encoded tensors. Gain practical insights on preparing and visualizing image datasets, creating minibatches, and optimizing your deep learning workflow for improved performance and efficiency.
Overview
Syllabus
A background on Fashion-MNIST
PyTorch's Dataset and DataLoaders usage example
Moving tensors to CUDA for GPU training
Transform input data with PyTorch's transform functions
Using target_transform for one hot encoded tensors lambda function
Taught by
Samuel Chan
Related Courses
-
PyTorch for Deep Learning & Machine Learning – Full Course
-
PyTorch Basics for Machine Learning
-
Advanced Generative Adversarial Networks (GANs)
-
Deep Learning - Computer Vision for Beginners Using PyTorch
-
Deep Learning with PyTorch: Zero to GANs
5.0 -
Introduction to Tensors - PyTorch with GPU Series Part 2
