Learn to build and train a PyTorch neural network for image classification in this comprehensive tutorial video. Begin with a mental model of the training process, then dive into PyTorch imports and hyperparameter setup. Explore how to prepare training and test data using the FashionMNIST dataset. Create a neural network architecture using PyTorch's nn module, and understand loss functions, cross-entropy loss, and optimizers. Set up a complete training loop for forward and backward passes, achieving 85% test accuracy on FashionMNIST. Visualize the computation graph using pytorchviz, and learn how to save and load PyTorch models. Perfect for those with basic PyTorch knowledge looking to deepen their understanding of neural network training and implementation.
Overview
Syllabus
Mental model: Training a neural network in PyTorch
PyTorch imports and hyperparemeters
Training and test data for our FashionMNIST classification
Creating our Neural Network using PyTorch's nn module
Loss function, Cross-Entropy Loss and Optimizers
Setting up a training loop for the forward pass + backward pass
Visualize our computation graph with pytorchviz
Saving and loading pytorch models
Taught by
Samuel Chan
Related Courses
-
Introduction to Deep Learning with PyTorch
-
Building a Neural Network in PyTorch
-
Deep Learning with PyTorch : Image Segmentation
-
PyTorch for Deep Learning & Machine Learning – Full Course
-
Practical Guide to Neural Network Training: Working with Leading Frameworks
-
Mastering Neural Networks and Model Regularization
