sg_execution_times

在 Google Colab 中运行

Colab

下载 Notebook

Notebook

在 GitHub 上查看

GitHub

计算时间

00:29.573 124个文件的总执行时间 来自所有图库

示例	时间	内存 (MB)
使用foreach_map和torch.compile进行显式水平融合 (recipes_source/foreach_map.py)	00:16.082	0.0
每个样本的梯度 (intermediate_source/per_sample_grads.py)	00:07.904	0.0
sphx_glr_beginner_basics_transforms_tutorial.py (beginner_source/basics/transforms_tutorial.py)	00:04.404	0.0
序列模型和长短期记忆网络 (beginner_source/nlp/sequence_models_tutorial.py)	00:00.847	0.0
前向模式自动微分 (Beta) (intermediate_source/forward_ad_usage.py)	00:00.122	0.0
模板教程 (beginner_source/template_tutorial.py)	00:00.006	0.0
聊天机器人教程 (beginner_source/chatbot_tutorial.py)	00:00.003	0.0
知识蒸馏教程 (beginner_source/knowledge_distillation_tutorial.py)	00:00.003	0.0
DCGAN教程 (beginner_source/dcgan_faces_tutorial.py)	00:00.003	0.0
PyTorch入门 (beginner_source/nlp/pytorch_tutorial.py)	00:00.003	0.0
(Beta) 使用缩放点积注意力 (SDPA) 实现高性能 Transformer (intermediate_source/scaled_dot_product_attention_tutorial.py)	00:00.003	0.0
使用PyTorch进行深度学习 (beginner_source/nlp/deep_learning_tutorial.py)	00:00.003	0.0
torch.export教程 (intermediate_source/torch_export_tutorial.py)	00:00.003	0.0
torch.compile入门 (intermediate_source/torch_compile_tutorial.py)	00:00.003	0.0
通过替换nn.Transformer与Nested Tensors和torch.compile加速PyTorch Transformer (intermediate_source/transformer_building_blocks.py)	00:00.003	0.0
sphx_glr_beginner_introyt_introyt1_tutorial.py (beginner_source/introyt/introyt1_tutorial.py)	00:00.003	0.0
Jacobians、Hessians、hvp、vhp等：组合函数变换 (intermediate_source/jacobians_hessians.py)	00:00.003	0.0
对抗性示例生成 (beginner_source/fgsm_tutorial.py)	00:00.003	0.0
sphx_glr_beginner_introyt_trainingyt.py (beginner_source/introyt/trainingyt.py)	00:00.003	0.0
训练一个分类器 (beginner_source/blitz/cifar10_tutorial.py)	00:00.003	0.0
Inductor CPU后端调试和分析 (intermediate_source/inductor_debug_cpu.py)	00:00.003	0.0
PyTorch中non_blocking和pin_memory()的良好使用指南 (intermediate_source/pinmem_nonblock.py)	00:00.003	0.0
计算机视觉迁移学习教程 (beginner_source/transfer_learning_tutorial.py)	00:00.003	0.0
sphx_glr_beginner_introyt_tensorboardyt_tutorial.py (beginner_source/introyt/tensorboardyt_tutorial.py)	00:00.003	0.0
从零开始的NLP：使用字符级RNN生成名称 (intermediate_source/char_rnn_generation_tutorial.py)	00:00.003	0.0
(prototype) 使用TorchAO进行GPU量化 (unstable_source/gpu_quantization_torchao_tutorial.py)	00:00.003	0.0
autograd保存张量的hooks (intermediate_source/autograd_saved_tensors_hooks_tutorial.py)	00:00.003	0.0
使用PyTorch进行神经风格迁移 (advanced_source/neural_style_tutorial.py)	00:00.003	0.0
TorchVision目标检测微调教程 (intermediate_source/torchvision_tutorial.py)	00:00.002	0.0
PyTorch中的Channels Last内存格式 (intermediate_source/memory_format_tutorial.py)	00:00.002	0.0
torch.compile端到端教程 (intermediate_source/torch_compile_full_example.py)	00:00.002	0.0
使用TorchRL进行强化学习 (PPO) 教程 (intermediate_source/reinforcement_ppo.py)	00:00.002	0.0
训练一个玩马里奥的强化学习智能体 (intermediate_source/mario_rl_tutorial.py)	00:00.002	0.0
TorchRL目标：编写DDPG损失函数 (advanced_source/coding_ddpg.py)	00:00.002	0.0
MaskedTensor概览 (unstable_source/maskedtensor_overview.py)	00:00.002	0.0
空间变换网络教程 (intermediate_source/spatial_transformer_tutorial.py)	00:00.002	0.0
使用MaskedTensor高效编写Adagrad的“稀疏”语义 (unstable_source/maskedtensor_adagrad.py)	00:00.002	0.0
编写自定义数据集、DataLoaders和Transforms (beginner_source/data_loading_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_onnx_onnx_registry_tutorial.py (beginner_source/onnx/onnx_registry_tutorial.py)	00:00.002	0.0
torch.vmap (unstable_source/vmap_recipe.py)	00:00.002	0.0
sphx_glr_beginner_basics_quickstart_tutorial.py (beginner_source/basics/quickstart_tutorial.py)	00:00.002	0.0
使用torch.compiler.set_stance控制动态编译 (recipes_source/torch_compiler_set_stance_tutorial.py)	00:00.002	0.0
理解requires_grad、retain_grad、Leaf和Non-leaf Tensors (beginner_source/understanding_leaf_vs_nonleaf_tutorial.py)	00:00.002	0.0
如何通过融合优化器步长到反向传播中节省内存 (intermediate_source/optimizer_step_in_backward_tutorial.py)	00:00.002	0.0
强化学习 (DQN) 教程 (intermediate_source/reinforcement_q_learning.py)	00:00.002	0.0
参数化教程 (intermediate_source/parametrizations.py)	00:00.002	0.0
张量 (beginner_source/blitz/tensor_tutorial.py)	00:00.002	0.0
进阶：做出动态决策和Bi-LSTM CRF (beginner_source/nlp/advanced_tutorial.py)	00:00.002	0.0
torch.autograd 简介 (beginner_source/blitz/autograd_tutorial.py)	00:00.002	0.0
(beta) 运行编译后的优化器和LR调度器 (recipes_source/compiling_optimizer_lr_scheduler.py)	00:00.002	0.0
神经切线核 (intermediate_source/neural_tangent_kernels.py)	00:00.002	0.0
模型集成 (intermediate_source/ensembling.py)	00:00.002	0.0
sphx_glr_beginner_introyt_tensors_deeper_tutorial.py (beginner_source/introyt/tensors_deeper_tutorial.py)	00:00.002	0.0
torch.nn 到底是什么？ (beginner_source/nn_tutorial.py)	00:00.002	0.0
从零开始的NLP：使用字符级RNN对名称进行分类 (intermediate_source/char_rnn_classification_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_basics_tensorqs_tutorial.py (beginner_source/basics/tensorqs_tutorial.py)	00:00.002	0.0
Nested Tensors入门 (unstable_source/nestedtensor.py)	00:00.002	0.0
自定义Python操作符 (advanced_source/python_custom_ops.py)	00:00.002	0.0
词嵌入：编码词汇语义 (beginner_source/nlp/word_embeddings_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_onnx_export_simple_model_to_onnx_tutorial.py (beginner_source/onnx/export_simple_model_to_onnx_tutorial.py)	00:00.002	0.0
从检查点加载nn.Module的技巧 (recipes_source/recipes/module_load_state_dict_tips.py)	00:00.002	0.0
从零开始的NLP：使用序列到序列网络和注意力进行翻译 (intermediate_source/seq2seq_translation_tutorial.py)	00:00.002	0.0
使用Ray Tune进行超参数调优 (beginner_source/hyperparameter_tuning_tutorial.py)	00:00.002	0.0
(beta) 使用TORCH_LOGS Python API与torch.compile (recipes_source/torch_logs.py)	00:00.002	0.0
MaskedTensor稀疏性 (unstable_source/maskedtensor_sparsity.py)	00:00.002	0.0
剪枝教程 (intermediate_source/pruning_tutorial.py)	00:00.002	0.0
Pendulum：使用TorchRL编写环境和Transforms (advanced_source/pendulum.py)	00:00.002	0.0
可视化梯度 (intermediate_source/visualizing_gradients_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_basics_data_tutorial.py (beginner_source/basics/data_tutorial.py)	00:00.002	0.0
可选：数据并行 (beginner_source/blitz/data_parallel_tutorial.py)	00:00.002	0.0
使用自定义函数融合卷积和批量归一化 (intermediate_source/custom_function_conv_bn_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_basics_optimization_tutorial.py (beginner_source/basics/optimization_tutorial.py)	00:00.002	0.0
MaskedTensor高级语义 (unstable_source/maskedtensor_advanced_semantics.py)	00:00.002	0.0
使用NumPy和SciPy创建扩展 (advanced_source/numpy_extensions_tutorial.py)	00:00.002	0.0
(beta) 利用Torch Function模式与torch.compile (recipes_source/torch_compile_torch_function_modes.py)	00:00.002	0.0
使用Ax进行多目标NAS (intermediate_source/ax_multiobjective_nas_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_introyt_autogradyt_tutorial.py (beginner_source/introyt/autogradyt_tutorial.py)	00:00.002	0.0
在PyTorch中推理形状 (recipes_source/recipes/reasoning_about_shapes.py)	00:00.002	0.0
使用区域编译减少torch.compile的冷启动编译时间 (recipes_source/regional_compilation.py)	00:00.002	0.0
nn.Module中用于load_state_dict和张量子类的扩展点 (recipes_source/recipes/swap_tensors.py)	00:00.002	0.0
PyTorch: nn (beginner_source/examples_nn/polynomial_nn.py)	00:00.002	0.0
torch.export AOTInductor教程（Python运行时 Beta） (recipes_source/torch_export_aoti_python.py)	00:00.002	0.0
sphx_glr_beginner_basics_autogradqs_tutorial.py (beginner_source/basics/autogradqs_tutorial.py)	00:00.002	0.0
使用区域编译减少AoT冷启动编译时间 (recipes_source/regional_aot.py)	00:00.002	0.0
sphx_glr_beginner_basics_saveloadrun_tutorial.py (beginner_source/basics/saveloadrun_tutorial.py)	00:00.002	0.0
将用户定义的Triton内核与torch.compile一起使用 (recipes_source/torch_compile_user_defined_triton_kernel_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_basics_buildmodel_tutorial.py (beginner_source/basics/buildmodel_tutorial.py)	00:00.002	0.0
(beta) 使用FX构建简单的CPU性能分析器 (intermediate_source/fx_profiling_tutorial.py)	00:00.002	0.0
热身：numpy (beginner_source/examples_tensor/polynomial_numpy.py)	00:00.002	0.0
神经网络 (beginner_source/blitz/neural_networks_tutorial.py)	00:00.002	0.0
sphx_glr_beginner_onnx_export_control_flow_model_to_onnx_tutorial.py (beginner_source/onnx/export_control_flow_model_to_onnx_tutorial.py)	00:00.002	0.0
PyTorch: 张量 (beginner_source/examples_tensor/polynomial_tensor.py)	00:00.002	0.0
sphx_glr_beginner_introyt_modelsyt_tutorial.py (beginner_source/introyt/modelsyt_tutorial.py)	00:00.002	0.0
更改默认设备 (recipes_source/recipes/changing_default_device.py)	00:00.002	0.0
(beta) 使用半结构化（2:4）稀疏性加速BERT (advanced_source/semi_structured_sparse.py)	00:00.000	0.0
使用基于PyTorch的USB进行半监督学习 (advanced_source/usb_semisup_learn.py)	00:00.000	0.0
sphx_glr_beginner_basics_intro.py (beginner_source/basics/intro.py)	00:00.000	0.0
PyTorch: 张量和autograd (beginner_source/examples_autograd/polynomial_autograd.py)	00:00.000	0.0
PyTorch: 定义新的autograd函数 (beginner_source/examples_autograd/polynomial_custom_function.py)	00:00.000	0.0
PyTorch: 控制流 + 权重共享 (beginner_source/examples_nn/dynamic_net.py)	00:00.000	0.0
PyTorch: 自定义nn模块 (beginner_source/examples_nn/polynomial_module.py)	00:00.000	0.0
PyTorch: optim (beginner_source/examples_nn/polynomial_optim.py)	00:00.000	0.0
sphx_glr_beginner_introyt_captumyt.py (beginner_source/introyt/captumyt.py)	00:00.000	0.0
sphx_glr_beginner_introyt_introyt_index.py (beginner_source/introyt/introyt_index.py)	00:00.000	0.0
sphx_glr_beginner_onnx_intro_onnx.py (beginner_source/onnx/intro_onnx.py)	00:00.000	0.0
分析你的PyTorch模块 (beginner_source/profiler.py)	00:00.000	0.0
保存和加载模型 (beginner_source/saving_loading_models.py)	00:00.000	0.0
循环DQN：训练循环策略 (intermediate_source/dqn_with_rnn_tutorial.py)	00:00.000	0.0
sphx_glr_intermediate_mnist_train_nas.py (intermediate_source/mnist_train_nas.py)	00:00.000	0.0
使用TensorBoard的PyTorch分析器 (intermediate_source/tensorboard_profiler_tutorial.py)	00:00.000	0.0
使用torch.compile构建卷积/批量归一化融合器 (intermediate_source/torch_compile_conv_bn_fuser.py)	00:00.000	0.0
TorchRec入门 (intermediate_source/torchrec_intro_tutorial.py)	00:00.000	0.0
使用Captum进行模型可解释性 (recipes_source/recipes/Captum_Recipe.py)	00:00.000	0.0
自动混合精度 (recipes_source/recipes/amp_recipe.py)	00:00.000	0.0
SyntaxError (recipes_source/recipes/benchmark.py)	00:00.000	0.0
在PyTorch中定义神经网络 (recipes_source/recipes/defining_a_neural_network.py)	00:00.000	0.0
PyTorch分析器 (recipes_source/recipes/profiler_recipe.py)	00:00.000	0.0
如何在PyTorch中使用TensorBoard (recipes_source/recipes/tensorboard_with_pytorch.py)	00:00.000	0.0
Timer快速入门 (recipes_source/recipes/timer_quick_start.py)	00:00.000	0.0
性能调优指南 (recipes_source/recipes/tuning_guide.py)	00:00.000	0.0
在PyTorch中使用来自不同模型的参数来预启动模型 (recipes_source/recipes/warmstarting_model_using_parameters_from_a_different_model.py)	00:00.000	0.0
PyTorch中的state_dict是什么？ (recipes_source/recipes/what_is_state_dict.py)	00:00.000	0.0
在PyTorch中将梯度清零 (recipes_source/recipes/zeroing_out_gradients.py)	00:00.000	0.0
(prototype) 使用GPUDirect Storage加速torch.save和torch.load (unstable_source/gpu_direct_storage.py)	00:00.000	0.0