Computation times#
00:26.477 total execution time for 129 files from all galleries:
Example |
Time |
Mem (MB) |
|---|---|---|
(beta) Utilizing Torch Function modes with torch.compile ( |
00:11.063 |
0.0 |
How to save memory by fusing the optimizer step into the backward pass ( |
00:09.173 |
0.0 |
sphx_glr_beginner_basics_transforms_tutorial.py ( |
00:04.312 |
0.0 |
Model ensembling ( |
00:00.756 |
0.0 |
Using Variable Length Attention in PyTorch ( |
00:00.464 |
0.0 |
Introduction to PyTorch ( |
00:00.362 |
0.0 |
torch.vmap ( |
00:00.129 |
0.0 |
Chatbot Tutorial ( |
00:00.003 |
0.0 |
Knowledge Distillation Tutorial ( |
00:00.003 |
0.0 |
DCGAN Tutorial ( |
00:00.003 |
0.0 |
(Beta) Implementing High-Performance Transformers with Scaled Dot Product Attention (SDPA) ( |
00:00.003 |
0.0 |
torch.export Tutorial ( |
00:00.003 |
0.0 |
Accelerating PyTorch Transformers by replacing nn.Transformer with Nested Tensors and torch.compile() ( |
00:00.003 |
0.0 |
sphx_glr_beginner_introyt_introyt1_tutorial.py ( |
00:00.003 |
0.0 |
Deep Learning with PyTorch ( |
00:00.003 |
0.0 |
NLP From Scratch: Generating Names with a Character-Level RNN ( |
00:00.003 |
0.0 |
Introduction to torch.compile ( |
00:00.003 |
0.0 |
Inductor CPU backend debugging and profiling ( |
00:00.003 |
0.0 |
A guide on good usage of non_blocking and pin_memory() in PyTorch ( |
00:00.003 |
0.0 |
sphx_glr_beginner_introyt_trainingyt.py ( |
00:00.003 |
0.0 |
sphx_glr_beginner_introyt_tensorboardyt_tutorial.py ( |
00:00.003 |
0.0 |
Forward-mode Automatic Differentiation (Beta) ( |
00:00.003 |
0.0 |
TorchVision Object Detection Finetuning Tutorial ( |
00:00.003 |
0.0 |
Jacobians, Hessians, hvp, vhp, and more: composing function transforms ( |
00:00.003 |
0.0 |
Transfer Learning for Computer Vision Tutorial ( |
00:00.003 |
0.0 |
Adversarial Example Generation ( |
00:00.003 |
0.0 |
(prototype) GPU Quantization with TorchAO ( |
00:00.003 |
0.0 |
Channels Last Memory Format in PyTorch ( |
00:00.002 |
0.0 |
Training a Classifier ( |
00:00.002 |
0.0 |
Neural Transfer Using PyTorch ( |
00:00.002 |
0.0 |
MaskedTensor Overview ( |
00:00.002 |
0.0 |
Train a Mario-playing RL Agent ( |
00:00.002 |
0.0 |
Sequence Models and Long Short-Term Memory Networks ( |
00:00.002 |
0.0 |
torch.compile End-to-End Tutorial ( |
00:00.002 |
0.0 |
Reinforcement Learning (PPO) with TorchRL Tutorial ( |
00:00.002 |
0.0 |
Dynamic Compilation Control with torch.compiler.set_stance ( |
00:00.002 |
0.0 |
Hooks for autograd saved tensors ( |
00:00.002 |
0.0 |
TorchRL objectives: Coding a DDPG loss ( |
00:00.002 |
0.0 |
Advanced: Making Dynamic Decisions and the Bi-LSTM CRF ( |
00:00.002 |
0.0 |
sphx_glr_beginner_onnx_onnx_registry_tutorial.py ( |
00:00.002 |
0.0 |
Writing Custom Datasets, DataLoaders and Transforms ( |
00:00.002 |
0.0 |
Distributed training at scale with PyTorch and Ray Train ( |
00:00.002 |
0.0 |
Understanding requires_grad, retain_grad, Leaf, and Non-leaf Tensors ( |
00:00.002 |
0.0 |
NLP From Scratch: Classifying Names with a Character-Level RNN ( |
00:00.002 |
0.0 |
Spatial Transformer Networks Tutorial ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_quickstart_tutorial.py ( |
00:00.002 |
0.0 |
Explicit horizontal fusion with foreach_map and torch.compile ( |
00:00.002 |
0.0 |
What is torch.nn really? ( |
00:00.002 |
0.0 |
sphx_glr_beginner_introyt_tensors_deeper_tutorial.py ( |
00:00.002 |
0.0 |
(beta) Running the compiled optimizer with an LR Scheduler ( |
00:00.002 |
0.0 |
Data Loading Optimization in PyTorch ( |
00:00.002 |
0.0 |
Custom Python Operators ( |
00:00.002 |
0.0 |
A Gentle Introduction to torch.autograd ( |
00:00.002 |
0.0 |
Efficiently writing “sparse” semantics for Adagrad with MaskedTensor ( |
00:00.002 |
0.0 |
Parametrizations Tutorial ( |
00:00.002 |
0.0 |
Tensors ( |
00:00.002 |
0.0 |
Fusing Convolution and Batch Norm using Custom Function ( |
00:00.002 |
0.0 |
Getting Started with Nested Tensors ( |
00:00.002 |
0.0 |
Creating Extensions Using NumPy and SciPy ( |
00:00.002 |
0.0 |
Neural Tangent Kernels ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_tensorqs_tutorial.py ( |
00:00.002 |
0.0 |
Reinforcement Learning (DQN) Tutorial ( |
00:00.002 |
0.0 |
Serve PyTorch models at scale with Ray Serve ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_data_tutorial.py ( |
00:00.002 |
0.0 |
sphx_glr_beginner_onnx_export_simple_model_to_onnx_tutorial.py ( |
00:00.002 |
0.0 |
Optional: Data Parallelism ( |
00:00.002 |
0.0 |
Per-sample-gradients ( |
00:00.002 |
0.0 |
Word Embeddings: Encoding Lexical Semantics ( |
00:00.002 |
0.0 |
NLP From Scratch: Translation with a Sequence to Sequence Network and Attention ( |
00:00.002 |
0.0 |
MaskedTensor Advanced Semantics ( |
00:00.002 |
0.0 |
(beta) Using TORCH_LOGS python API with torch.compile ( |
00:00.002 |
0.0 |
Pendulum: Writing your environment and transforms with TorchRL ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_optimization_tutorial.py ( |
00:00.002 |
0.0 |
Tips for Loading an nn.Module from a Checkpoint ( |
00:00.002 |
0.0 |
Multi-Objective NAS with Ax ( |
00:00.002 |
0.0 |
MaskedTensor Sparsity ( |
00:00.002 |
0.0 |
Pruning Tutorial ( |
00:00.002 |
0.0 |
sphx_glr_beginner_introyt_autogradyt_tutorial.py ( |
00:00.002 |
0.0 |
Visualizing Gradients ( |
00:00.002 |
0.0 |
Template Tutorial ( |
00:00.002 |
0.0 |
Hyperparameter tuning using Ray Tune ( |
00:00.002 |
0.0 |
Reducing torch.compile cold start compilation time with regional compilation ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_buildmodel_tutorial.py ( |
00:00.002 |
0.0 |
Using User-Defined Triton Kernels with torch.compile ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_autogradqs_tutorial.py ( |
00:00.002 |
0.0 |
(beta) Building a Simple CPU Performance Profiler with FX ( |
00:00.002 |
0.0 |
Reasoning about Shapes in PyTorch ( |
00:00.002 |
0.0 |
Neural Networks ( |
00:00.002 |
0.0 |
Extension points in nn.Module for load_state_dict and tensor subclasses ( |
00:00.002 |
0.0 |
PyTorch: nn ( |
00:00.002 |
0.0 |
Warm-up: numpy ( |
00:00.002 |
0.0 |
Changing default device ( |
00:00.002 |
0.0 |
Reducing AoT cold start compilation time with regional compilation ( |
00:00.002 |
0.0 |
sphx_glr_beginner_onnx_export_control_flow_model_to_onnx_tutorial.py ( |
00:00.002 |
0.0 |
torch.export AOTInductor Tutorial for Python runtime (Beta) ( |
00:00.002 |
0.0 |
DebugMode: Recording Dispatched Operations and Numerical Debugging ( |
00:00.002 |
0.0 |
sphx_glr_beginner_basics_saveloadrun_tutorial.py ( |
00:00.002 |
0.0 |
sphx_glr_beginner_introyt_modelsyt_tutorial.py ( |
00:00.002 |
0.0 |
PyTorch: Tensors ( |
00:00.002 |
0.0 |
(beta) Accelerating BERT with semi-structured (2:4) sparsity ( |
00:00.000 |
0.0 |
Semi-Supervised Learning using USB built upon PyTorch ( |
00:00.000 |
0.0 |
sphx_glr_beginner_basics_intro.py ( |
00:00.000 |
0.0 |
PyTorch: Tensors and autograd ( |
00:00.000 |
0.0 |
PyTorch: Defining New autograd Functions ( |
00:00.000 |
0.0 |
PyTorch: Control Flow + Weight Sharing ( |
00:00.000 |
0.0 |
PyTorch: Custom nn Modules ( |
00:00.000 |
0.0 |
PyTorch: optim ( |
00:00.000 |
0.0 |
sphx_glr_beginner_introyt_captumyt.py ( |
00:00.000 |
0.0 |
sphx_glr_beginner_introyt_introyt_index.py ( |
00:00.000 |
0.0 |
Mosaic: Memory Profiling for PyTorch ( |
00:00.000 |
0.0 |
sphx_glr_beginner_onnx_intro_onnx.py ( |
00:00.000 |
0.0 |
Profiling your PyTorch Module ( |
00:00.000 |
0.0 |
Saving and Loading Models ( |
00:00.000 |
0.0 |
Recurrent DQN: Training recurrent policies ( |
00:00.000 |
0.0 |
sphx_glr_intermediate_mnist_train_nas.py ( |
00:00.000 |
0.0 |
Building a Convolution/Batch Norm fuser with torch.compile ( |
00:00.000 |
0.0 |
Introduction to TorchRec ( |
00:00.000 |
0.0 |
Model Interpretability using Captum ( |
00:00.000 |
0.0 |
Automatic Mixed Precision ( |
00:00.000 |
0.0 |
SyntaxError ( |
00:00.000 |
0.0 |
Defining a Neural Network in PyTorch ( |
00:00.000 |
0.0 |
PyTorch Profiler ( |
00:00.000 |
0.0 |
How to use TensorBoard with PyTorch ( |
00:00.000 |
0.0 |
Timer quick start ( |
00:00.000 |
0.0 |
Performance Tuning Guide ( |
00:00.000 |
0.0 |
Warmstarting model using parameters from a different model in PyTorch ( |
00:00.000 |
0.0 |
What is a state_dict in PyTorch ( |
00:00.000 |
0.0 |
Zeroing out gradients in PyTorch ( |
00:00.000 |
0.0 |
(prototype) Accelerating torch.save and torch.load with GPUDirect Storage ( |
00:00.000 |
0.0 |