Deep Learning

12 pages in this section.

Why tensors, autograd, the training loop, and GPU/mixed-precision training are one machine rather than four separate topics - the mental model behind every other page in this section.

Explainer Python Deep Learning Autograd

PyTorch Basics

Tensors, devices, and the core API.

Python Deep Learning Pytorch Basics

Autograd

Automatic differentiation and computation graphs.

Python Deep Learning Autograd Gradients

Building Models with nn.Module

Layers, forward passes, and parameters.

Python Deep Learning Nn Module Models

Datasets & DataLoaders

Efficient input pipelines and transforms.

Python Deep Learning Dataloader Datasets

Training Loops

Loss, optimizers, schedulers, and checkpoints.

Python Deep Learning Training Optimizers

GPUs & Mixed Precision

Python Deep Learning Gpu Mixed Precision

Transfer Learning & Fine-Tuning

Python Deep Learning Transfer Learning Fine Tuning

PyTorch Lightning

Python Deep Learning Lightning Training

Saving, Loading & Export

Python Deep Learning Export Onnx

Distributed Training

Python Deep Learning Distributed Ddp

Deep Learning Best Practices

Reproducibility and efficient training.

Python Deep Learning Best Practices List

Deep Learning

Search Documentation

Deep Learning