The Thunder Sessions | Session 21 | How the DeepSeek R1 Sausage is made

The Thunder Sessions | Session 20

The Thunder Sessions | Session 19

The Thunder Sessions | Session 18

The Thunder Sessions | Session 17

The Thunder Sessions | Session 16

The Thunder Sessions | Session 15

The Thunder Sessions | Session 14

The Thunder Sessions | Session 13

More GPU Machines Now 50-80 Percent Cheaper

The Thunder Sessions | Session 12

Thunder Sessions | Session 11

The Thunder Sessions | Session 10 | LLL: Liger-Kernel+LitGPT+Llama 3.2

The Thunder Sessions | Session 9 | Applying custom kernels and fusions to large models

The Thunder Sessions | Session 8

The Thunder Sessions | Session 7 | Fusing Kernels with Thunder & Triton

The Thunder Sessions | Session 6 | More Transforms, Less Theory

Meet LitServe - The fast, simple way to deploy AI models

The Thunder Sessions | Session 5 | Speedups in Llama 3

The Thunder Sessions | Session 2

The Thunder Sessions | Session 4 | Transforms

The Thunder Sesssions | Session 3 | Deep Learning Compilers

The Thunder Sessions | Session 1

Meet Studio Templates | Get started today for free with Studio Templates

Meet Studio | Turn ideas into AI, Lightning Fast | from Lightning AI Creators of PyTorch Lightning

AI Regulation: A Fireside Conversation about AI's relationship with Washington DC

Implementing Your AI Strategy with Lightning Studio | Luca Antiga presentation at AI Summit Dec 2023

Unit 2.2 | What are Tensors? | Part 01 | Tensors for Data

Unit 4.1 | Logistic Regression for Multiple Classes | Part 2 | The Softmax Activation Function

Unit 4.1 | Logistic Regression for Multiple Classes | Part 5 | The Cross Entropy Loss Function

Unit 4.1 | Logistic Regression for Multiple Classes | Part 4 | Cross Entropy Loss Function

Unit 4.2 | Multilayer Neural Networks | Part 1 | Looking Beyond Linear Decision Boundaries

Unit 4.2 | Multilayer Neural Networks | Part 2 | The Multilayer Perceptron Architecture

Unit 4.2 | Multilayer Neural Networks | Part 3 | Basic Architecture Design Considerations

Unit 4.3 | Training a Multilayer Perceptron in PyTorch | Part 1

Unit 4.3 | Training a Multilayer Perceptron in PyTorch | Part 3

Unit 4.3 | Training a Multilayer Perceptron in PyTorch | Part 4

Unit 4.3 | Training a Multilayer Perceptron in PyTorch | Part 5

Unit 4.4 | Defining Efficient Data Loaders | Part 1 | Avoiding Data Loading Bottlenecks

Unit 4.4 | Defining Efficient Data Loaders | Part 2 | Datasets and Dataloaders

Unit 4.4 | Defining Efficient Data Loaders | Part 3 | Coding

Unit 4.4 | Defining Efficient Data Loaders | Part 4 | Coding

Unit 4.6 | Speeding Up Model Training Using GPUs

Unit 4.5 | Multilayer Neural Networks for Regression | Part 2 | Coding

Unit 5 | Organizing Your PyTorch Code with Lightning

Unit 5.1 | Getting Started with Structuring Your PyTorch Code using Lightning

Unit 5.2 | Training a Multilayer Perceptron in PyTorch Lightning | Part 1

Unit 5.2 | Training a Multilayer Perceptron in PyTorch Lightning | Part 2

Unit 5.3 | Computing Metrics Efficiently with TorchMetrics | Part 1 | How Does TorchMetrics Work?

Unit 5.3 | Computing Metrics Efficiently with TorchMetrics | Part 2

Unit 5.3 | Computing Metrics Efficiently with TorchMetrics | Part 3

Unit 5.4 | Making Code Reproducible | Part 2 | Controlling Sources of Randomness

Unit 5.4 | Making Code Reproducible | Part 3 | Coding

Unit 5.5 | Organizing Your Data Loaders with Data Modules | Part 1

Unit 5.5 | Organizing Your Data Loaders with Data Modules | Part 2

Unit 5.6 | The Benefits of Logging | Part 1 | Tracking Training Progress

Unit 5.7 | Evaluating and Using Models on New Data | Part 1 | Fundamental Model Inspection Steps

Unit 5.6 | The Benefits of Logging | Part 2 | Coding

Unit 6.1 | Model Checkpointing and Early Stopping | Part 1

Unit 6.1 | Model Checkpointing and Early Stopping | Part 3