├── LICENSE ├── README.md ├── assets ├── HTM.gif └── VTM.gif ├── env.yaml ├── experiments ├── apex_copy │ ├── .github │ │ └── ISSUE_TEMPLATE │ │ │ └── bug_report.md │ ├── .gitmodules │ ├── .nojekyll │ ├── LICENSE │ ├── README.md │ ├── apex.egg-info │ │ ├── PKG-INFO │ │ ├── SOURCES.txt │ │ ├── dependency_links.txt │ │ └── top_level.txt │ ├── apex │ │ ├── RNN │ │ │ ├── README.md │ │ │ ├── RNNBackend.py │ │ │ ├── __init__.py │ │ │ ├── cells.py │ │ │ └── models.py │ │ ├── __init__.py │ │ ├── _autocast_utils.py │ │ ├── amp │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── __version__.py │ │ │ ├── _amp_state.py │ │ │ ├── _initialize.py │ │ │ ├── _process_optimizer.py │ │ │ ├── amp.py │ │ │ ├── compat.py │ │ │ ├── frontend.py │ │ │ ├── handle.py │ │ │ ├── lists │ │ │ │ ├── __init__.py │ │ │ │ ├── functional_overrides.py │ │ │ │ ├── tensor_overrides.py │ │ │ │ └── torch_overrides.py │ │ │ ├── opt.py │ │ │ ├── rnn_compat.py │ │ │ ├── scaler.py │ │ │ ├── utils.py │ │ │ └── wrap.py │ │ ├── contrib │ │ │ ├── __init__.py │ │ │ ├── bottleneck │ │ │ │ ├── __init__.py │ │ │ │ ├── bottleneck.py │ │ │ │ ├── halo_exchangers.py │ │ │ │ └── test.py │ │ │ ├── clip_grad │ │ │ │ ├── __init__.py │ │ │ │ └── clip_grad.py │ │ │ ├── conv_bias_relu │ │ │ │ ├── __init__.py │ │ │ │ └── conv_bias_relu.py │ │ │ ├── csrc │ │ │ │ ├── bottleneck │ │ │ │ │ └── bottleneck.cpp │ │ │ │ ├── conv_bias_relu │ │ │ │ │ └── conv_bias_relu.cpp │ │ │ │ ├── cudnn_gbn │ │ │ │ │ ├── cudnn_gbn.cpp │ │ │ │ │ ├── norm_sample.cpp │ │ │ │ │ └── norm_sample.h │ │ │ │ ├── fmha │ │ │ │ │ ├── fmha_api.cpp │ │ │ │ │ └── src │ │ │ │ │ │ ├── fmha.h │ │ │ │ │ │ ├── fmha │ │ │ │ │ │ ├── gemm.h │ │ │ │ │ │ ├── gmem_tile.h │ │ │ │ │ │ ├── kernel_traits.h │ │ │ │ │ │ ├── mask.h │ │ │ │ │ │ ├── smem_tile.h │ │ │ │ │ │ ├── softmax.h │ │ │ │ │ │ └── utils.h │ │ │ │ │ │ ├── fmha_dgrad_fp16_128_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_dgrad_fp16_256_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_dgrad_fp16_384_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_dgrad_fp16_512_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_dgrad_kernel_1xN_reload.h │ │ │ │ │ │ ├── fmha_dgrad_kernel_1xN_reload_nl.h │ │ │ │ │ │ ├── fmha_fprop_fp16_128_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_fprop_fp16_256_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_fprop_fp16_384_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_fprop_fp16_512_64_kernel.sm80.cu │ │ │ │ │ │ ├── fmha_fprop_kernel_1xN.h │ │ │ │ │ │ ├── fmha_kernel.h │ │ │ │ │ │ ├── fmha_noloop_reduce.cu │ │ │ │ │ │ └── fmha_utils.h │ │ │ │ ├── focal_loss │ │ │ │ │ ├── focal_loss_cuda.cpp │ │ │ │ │ └── focal_loss_cuda_kernel.cu │ │ │ │ ├── groupbn │ │ │ │ │ ├── batch_norm.cu │ │ │ │ │ ├── batch_norm.h │ │ │ │ │ ├── batch_norm_add_relu.cu │ │ │ │ │ ├── batch_norm_add_relu.h │ │ │ │ │ ├── cuda_utils.h │ │ │ │ │ ├── interface.cpp │ │ │ │ │ ├── ipc.cu │ │ │ │ │ └── nhwc_batch_norm_kernel.h │ │ │ │ ├── index_mul_2d │ │ │ │ │ ├── index_mul_2d_cuda.cpp │ │ │ │ │ └── index_mul_2d_cuda_kernel.cu │ │ │ │ ├── layer_norm │ │ │ │ │ ├── ln.h │ │ │ │ │ ├── ln_api.cpp │ │ │ │ │ ├── ln_bwd_kernels.cuh │ │ │ │ │ ├── ln_bwd_semi_cuda_kernel.cu │ │ │ │ │ ├── ln_fwd_cuda_kernel.cu │ │ │ │ │ ├── ln_fwd_kernels.cuh │ │ │ │ │ ├── ln_kernel_traits.h │ │ │ │ │ └── ln_utils.cuh │ │ │ │ ├── multihead_attn │ │ │ │ │ ├── additive_masked_softmax_dropout_cuda.cu │ │ │ │ │ ├── dropout.cuh │ │ │ │ │ ├── encdec_multihead_attn_cuda.cu │ │ │ │ │ ├── encdec_multihead_attn_norm_add_cuda.cu │ │ │ │ │ ├── layer_norm.cuh │ │ │ │ │ ├── masked_softmax_dropout_cuda.cu │ │ │ │ │ ├── multihead_attn_frontend.cpp │ │ │ │ │ ├── philox.cuh │ │ │ │ │ ├── self_multihead_attn_bias_additive_mask_cuda.cu │ │ │ │ │ ├── self_multihead_attn_bias_cuda.cu │ │ │ │ │ ├── self_multihead_attn_cuda.cu │ │ │ │ │ ├── self_multihead_attn_norm_add_cuda.cu │ │ │ │ │ ├── softmax.cuh │ │ │ │ │ └── strided_batched_gemm.cuh │ │ │ │ ├── nccl_p2p │ │ │ │ │ ├── nccl_p2p.cpp │ │ │ │ │ ├── nccl_p2p_cuda.cu │ │ │ │ │ ├── nccl_p2p_cuda.cuh │ │ │ │ │ ├── nccl_version.cpp │ │ │ │ │ └── nccl_version_check.cu │ │ │ │ ├── optimizers │ │ │ │ │ ├── fused_adam_cuda.cpp │ │ │ │ │ ├── fused_adam_cuda_kernel.cu │ │ │ │ │ ├── fused_lamb_cuda.cpp │ │ │ │ │ ├── fused_lamb_cuda_kernel.cu │ │ │ │ │ ├── multi_tensor_distopt_adam.cpp │ │ │ │ │ ├── multi_tensor_distopt_adam_kernel.cu │ │ │ │ │ ├── multi_tensor_distopt_lamb.cpp │ │ │ │ │ └── multi_tensor_distopt_lamb_kernel.cu │ │ │ │ ├── peer_memory │ │ │ │ │ ├── peer_memory.cpp │ │ │ │ │ ├── peer_memory_cuda.cu │ │ │ │ │ └── peer_memory_cuda.cuh │ │ │ │ ├── transducer │ │ │ │ │ ├── transducer_joint.cpp │ │ │ │ │ ├── transducer_joint_kernel.cu │ │ │ │ │ ├── transducer_loss.cpp │ │ │ │ │ └── transducer_loss_kernel.cu │ │ │ │ └── xentropy │ │ │ │ │ ├── interface.cpp │ │ │ │ │ └── xentropy_kernel.cu │ │ │ ├── cudnn_gbn │ │ │ │ ├── __init__.py │ │ │ │ └── batch_norm.py │ │ │ ├── examples │ │ │ │ └── multihead_attn │ │ │ │ │ ├── func_test_multihead_attn.py │ │ │ │ │ └── perf_test_multihead_attn.py │ │ │ ├── fmha │ │ │ │ ├── __init__.py │ │ │ │ └── fmha.py │ │ │ ├── focal_loss │ │ │ │ ├── __init__.py │ │ │ │ └── focal_loss.py │ │ │ ├── groupbn │ │ │ │ ├── __init__.py │ │ │ │ └── batch_norm.py │ │ │ ├── index_mul_2d │ │ │ │ ├── __init__.py │ │ │ │ └── index_mul_2d.py │ │ │ ├── layer_norm │ │ │ │ ├── __init__.py │ │ │ │ └── layer_norm.py │ │ │ ├── multihead_attn │ │ │ │ ├── MHA_bwd.png │ │ │ │ ├── MHA_fwd.png │ │ │ │ ├── README.md │ │ │ │ ├── __init__.py │ │ │ │ ├── encdec_multihead_attn.py │ │ │ │ ├── encdec_multihead_attn_func.py │ │ │ │ ├── fast_encdec_multihead_attn_func.py │ │ │ │ ├── fast_encdec_multihead_attn_norm_add_func.py │ │ │ │ ├── fast_self_multihead_attn_func.py │ │ │ │ ├── fast_self_multihead_attn_norm_add_func.py │ │ │ │ ├── mask_softmax_dropout_func.py │ │ │ │ ├── self_multihead_attn.py │ │ │ │ └── self_multihead_attn_func.py │ │ │ ├── optimizers │ │ │ │ ├── __init__.py │ │ │ │ ├── distributed_fused_adam.py │ │ │ │ ├── distributed_fused_lamb.py │ │ │ │ ├── fp16_optimizer.py │ │ │ │ ├── fused_adam.py │ │ │ │ ├── fused_lamb.py │ │ │ │ └── fused_sgd.py │ │ │ ├── peer_memory │ │ │ │ ├── __init__.py │ │ │ │ ├── peer_halo_exchanger_1d.py │ │ │ │ └── peer_memory.py │ │ │ ├── sparsity │ │ │ │ ├── COPYRIGHT │ │ │ │ ├── README.md │ │ │ │ ├── __init__.py │ │ │ │ ├── asp.py │ │ │ │ ├── permutation_lib.py │ │ │ │ ├── permutation_search_kernels │ │ │ │ │ ├── CUDA_kernels │ │ │ │ │ │ └── permutation_search_kernels.cu │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── call_permutation_search_kernels.py │ │ │ │ │ ├── channel_swap.py │ │ │ │ │ ├── exhaustive_search.py │ │ │ │ │ └── permutation_utilities.py │ │ │ │ ├── permutation_tests │ │ │ │ │ ├── README.md │ │ │ │ │ ├── ablation_studies.sh │ │ │ │ │ ├── permutation_test.py │ │ │ │ │ ├── runtime_table.sh │ │ │ │ │ └── unstructured_study.sh │ │ │ │ ├── sparse_masklib.py │ │ │ │ └── test │ │ │ │ │ ├── checkpointing_test_part1.py │ │ │ │ │ ├── checkpointing_test_part2.py │ │ │ │ │ ├── checkpointing_test_reference.py │ │ │ │ │ └── toy_problem.py │ │ │ ├── test │ │ │ │ ├── __init__.py │ │ │ │ ├── bottleneck │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_bottleneck_module.py │ │ │ │ ├── clip_grad │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_clip_grad.py │ │ │ │ ├── conv_bias_relu │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_conv_bias_relu.py │ │ │ │ ├── cudnn_gbn │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_cudnn_gbn_with_two_gpus.py │ │ │ │ ├── fmha │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_fmha.py │ │ │ │ ├── focal_loss │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_focal_loss.py │ │ │ │ ├── fused_dense │ │ │ │ │ └── test_fused_dense.py │ │ │ │ ├── index_mul_2d │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_index_mul_2d.py │ │ │ │ ├── layer_norm │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_fast_layer_norm.py │ │ │ │ ├── multihead_attn │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── test_encdec_multihead_attn.py │ │ │ │ │ ├── test_encdec_multihead_attn_norm_add.py │ │ │ │ │ ├── test_fast_self_multihead_attn_bias.py │ │ │ │ │ ├── test_mha_fused_softmax.py │ │ │ │ │ ├── test_self_multihead_attn.py │ │ │ │ │ └── test_self_multihead_attn_norm_add.py │ │ │ │ ├── optimizers │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_dist_adam.py │ │ │ │ ├── peer_memory │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_peer_halo_exchange_module.py │ │ │ │ ├── transducer │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── test_transducer_joint.py │ │ │ │ │ └── test_transducer_loss.py │ │ │ │ └── xentropy │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_label_smoothing.py │ │ │ ├── transducer │ │ │ │ ├── __init__.py │ │ │ │ ├── _transducer_ref.py │ │ │ │ └── transducer.py │ │ │ └── xentropy │ │ │ │ ├── __init__.py │ │ │ │ └── softmax_xentropy.py │ │ ├── fp16_utils │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── fp16_optimizer.py │ │ │ ├── fp16util.py │ │ │ └── loss_scaler.py │ │ ├── fused_dense │ │ │ ├── __init__.py │ │ │ └── fused_dense.py │ │ ├── mlp │ │ │ ├── __init__.py │ │ │ └── mlp.py │ │ ├── multi_tensor_apply │ │ │ ├── __init__.py │ │ │ └── multi_tensor_apply.py │ │ ├── normalization │ │ │ ├── __init__.py │ │ │ └── fused_layer_norm.py │ │ ├── optimizers │ │ │ ├── __init__.py │ │ │ ├── fused_adagrad.py │ │ │ ├── fused_adam.py │ │ │ ├── fused_lamb.py │ │ │ ├── fused_mixed_precision_lamb.py │ │ │ ├── fused_novograd.py │ │ │ └── fused_sgd.py │ │ ├── parallel │ │ │ ├── LARC.py │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── distributed.py │ │ │ ├── multiproc.py │ │ │ ├── optimized_sync_batchnorm.py │ │ │ ├── optimized_sync_batchnorm_kernel.py │ │ │ ├── sync_batchnorm.py │ │ │ └── sync_batchnorm_kernel.py │ │ └── transformer │ │ │ ├── README.md │ │ │ ├── __init__.py │ │ │ ├── _data │ │ │ ├── __init__.py │ │ │ └── _batchsampler.py │ │ │ ├── amp │ │ │ ├── __init__.py │ │ │ └── grad_scaler.py │ │ │ ├── enums.py │ │ │ ├── functional │ │ │ ├── __init__.py │ │ │ └── fused_softmax.py │ │ │ ├── layers │ │ │ ├── __init__.py │ │ │ └── layer_norm.py │ │ │ ├── log_util.py │ │ │ ├── microbatches.py │ │ │ ├── parallel_state.py │ │ │ ├── pipeline_parallel │ │ │ ├── __init__.py │ │ │ ├── _timers.py │ │ │ ├── p2p_communication.py │ │ │ ├── schedules │ │ │ │ ├── __init__.py │ │ │ │ ├── common.py │ │ │ │ ├── fwd_bwd_no_pipelining.py │ │ │ │ ├── fwd_bwd_pipelining_with_interleaving.py │ │ │ │ └── fwd_bwd_pipelining_without_interleaving.py │ │ │ └── utils.py │ │ │ ├── tensor_parallel │ │ │ ├── __init__.py │ │ │ ├── cross_entropy.py │ │ │ ├── data.py │ │ │ ├── layers.py │ │ │ ├── mappings.py │ │ │ ├── memory.py │ │ │ ├── random.py │ │ │ └── utils.py │ │ │ ├── testing │ │ │ ├── __init__.py │ │ │ ├── arguments.py │ │ │ ├── commons.py │ │ │ ├── distributed_test_base.py │ │ │ ├── global_vars.py │ │ │ ├── standalone_bert.py │ │ │ ├── standalone_gpt.py │ │ │ └── standalone_transformer_lm.py │ │ │ └── utils.py │ ├── build │ │ └── lib │ │ │ └── apex │ │ │ ├── RNN │ │ │ ├── RNNBackend.py │ │ │ ├── __init__.py │ │ │ ├── cells.py │ │ │ └── models.py │ │ │ ├── __init__.py │ │ │ ├── _autocast_utils.py │ │ │ ├── amp │ │ │ ├── __init__.py │ │ │ ├── __version__.py │ │ │ ├── _amp_state.py │ │ │ ├── _initialize.py │ │ │ ├── _process_optimizer.py │ │ │ ├── amp.py │ │ │ ├── compat.py │ │ │ ├── frontend.py │ │ │ ├── handle.py │ │ │ ├── lists │ │ │ │ ├── __init__.py │ │ │ │ ├── functional_overrides.py │ │ │ │ ├── tensor_overrides.py │ │ │ │ └── torch_overrides.py │ │ │ ├── opt.py │ │ │ ├── rnn_compat.py │ │ │ ├── scaler.py │ │ │ ├── utils.py │ │ │ └── wrap.py │ │ │ ├── contrib │ │ │ ├── __init__.py │ │ │ ├── bottleneck │ │ │ │ ├── __init__.py │ │ │ │ ├── bottleneck.py │ │ │ │ ├── halo_exchangers.py │ │ │ │ └── test.py │ │ │ ├── clip_grad │ │ │ │ ├── __init__.py │ │ │ │ └── clip_grad.py │ │ │ ├── conv_bias_relu │ │ │ │ ├── __init__.py │ │ │ │ └── conv_bias_relu.py │ │ │ ├── cudnn_gbn │ │ │ │ ├── __init__.py │ │ │ │ └── batch_norm.py │ │ │ ├── fmha │ │ │ │ ├── __init__.py │ │ │ │ └── fmha.py │ │ │ ├── focal_loss │ │ │ │ ├── __init__.py │ │ │ │ └── focal_loss.py │ │ │ ├── groupbn │ │ │ │ ├── __init__.py │ │ │ │ └── batch_norm.py │ │ │ ├── index_mul_2d │ │ │ │ ├── __init__.py │ │ │ │ └── index_mul_2d.py │ │ │ ├── layer_norm │ │ │ │ ├── __init__.py │ │ │ │ └── layer_norm.py │ │ │ ├── multihead_attn │ │ │ │ ├── __init__.py │ │ │ │ ├── encdec_multihead_attn.py │ │ │ │ ├── encdec_multihead_attn_func.py │ │ │ │ ├── fast_encdec_multihead_attn_func.py │ │ │ │ ├── fast_encdec_multihead_attn_norm_add_func.py │ │ │ │ ├── fast_self_multihead_attn_func.py │ │ │ │ ├── fast_self_multihead_attn_norm_add_func.py │ │ │ │ ├── mask_softmax_dropout_func.py │ │ │ │ ├── self_multihead_attn.py │ │ │ │ └── self_multihead_attn_func.py │ │ │ ├── optimizers │ │ │ │ ├── __init__.py │ │ │ │ ├── distributed_fused_adam.py │ │ │ │ ├── distributed_fused_lamb.py │ │ │ │ ├── fp16_optimizer.py │ │ │ │ ├── fused_adam.py │ │ │ │ ├── fused_lamb.py │ │ │ │ └── fused_sgd.py │ │ │ ├── peer_memory │ │ │ │ ├── __init__.py │ │ │ │ ├── peer_halo_exchanger_1d.py │ │ │ │ └── peer_memory.py │ │ │ ├── sparsity │ │ │ │ ├── __init__.py │ │ │ │ ├── asp.py │ │ │ │ ├── permutation_lib.py │ │ │ │ ├── permutation_search_kernels │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── call_permutation_search_kernels.py │ │ │ │ │ ├── channel_swap.py │ │ │ │ │ ├── exhaustive_search.py │ │ │ │ │ └── permutation_utilities.py │ │ │ │ └── sparse_masklib.py │ │ │ ├── test │ │ │ │ ├── __init__.py │ │ │ │ ├── bottleneck │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_bottleneck_module.py │ │ │ │ ├── clip_grad │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_clip_grad.py │ │ │ │ ├── conv_bias_relu │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_conv_bias_relu.py │ │ │ │ ├── cudnn_gbn │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_cudnn_gbn_with_two_gpus.py │ │ │ │ ├── fmha │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_fmha.py │ │ │ │ ├── focal_loss │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_focal_loss.py │ │ │ │ ├── index_mul_2d │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_index_mul_2d.py │ │ │ │ ├── layer_norm │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_fast_layer_norm.py │ │ │ │ ├── multihead_attn │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── test_encdec_multihead_attn.py │ │ │ │ │ ├── test_encdec_multihead_attn_norm_add.py │ │ │ │ │ ├── test_fast_self_multihead_attn_bias.py │ │ │ │ │ ├── test_mha_fused_softmax.py │ │ │ │ │ ├── test_self_multihead_attn.py │ │ │ │ │ └── test_self_multihead_attn_norm_add.py │ │ │ │ ├── optimizers │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_dist_adam.py │ │ │ │ ├── peer_memory │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_peer_halo_exchange_module.py │ │ │ │ ├── transducer │ │ │ │ │ ├── __init__.py │ │ │ │ │ ├── test_transducer_joint.py │ │ │ │ │ └── test_transducer_loss.py │ │ │ │ └── xentropy │ │ │ │ │ ├── __init__.py │ │ │ │ │ └── test_label_smoothing.py │ │ │ ├── transducer │ │ │ │ ├── __init__.py │ │ │ │ ├── _transducer_ref.py │ │ │ │ └── transducer.py │ │ │ └── xentropy │ │ │ │ ├── __init__.py │ │ │ │ └── softmax_xentropy.py │ │ │ ├── fp16_utils │ │ │ ├── __init__.py │ │ │ ├── fp16_optimizer.py │ │ │ ├── fp16util.py │ │ │ └── loss_scaler.py │ │ │ ├── fused_dense │ │ │ ├── __init__.py │ │ │ └── fused_dense.py │ │ │ ├── mlp │ │ │ ├── __init__.py │ │ │ └── mlp.py │ │ │ ├── multi_tensor_apply │ │ │ ├── __init__.py │ │ │ └── multi_tensor_apply.py │ │ │ ├── normalization │ │ │ ├── __init__.py │ │ │ └── fused_layer_norm.py │ │ │ ├── optimizers │ │ │ ├── __init__.py │ │ │ ├── fused_adagrad.py │ │ │ ├── fused_adam.py │ │ │ ├── fused_lamb.py │ │ │ ├── fused_mixed_precision_lamb.py │ │ │ ├── fused_novograd.py │ │ │ └── fused_sgd.py │ │ │ ├── parallel │ │ │ ├── LARC.py │ │ │ ├── __init__.py │ │ │ ├── distributed.py │ │ │ ├── multiproc.py │ │ │ ├── optimized_sync_batchnorm.py │ │ │ ├── optimized_sync_batchnorm_kernel.py │ │ │ ├── sync_batchnorm.py │ │ │ └── sync_batchnorm_kernel.py │ │ │ └── transformer │ │ │ ├── __init__.py │ │ │ ├── _data │ │ │ ├── __init__.py │ │ │ └── _batchsampler.py │ │ │ ├── amp │ │ │ ├── __init__.py │ │ │ └── grad_scaler.py │ │ │ ├── enums.py │ │ │ ├── functional │ │ │ ├── __init__.py │ │ │ └── fused_softmax.py │ │ │ ├── layers │ │ │ ├── __init__.py │ │ │ └── layer_norm.py │ │ │ ├── log_util.py │ │ │ ├── microbatches.py │ │ │ ├── parallel_state.py │ │ │ ├── pipeline_parallel │ │ │ ├── __init__.py │ │ │ ├── _timers.py │ │ │ ├── p2p_communication.py │ │ │ ├── schedules │ │ │ │ ├── __init__.py │ │ │ │ ├── common.py │ │ │ │ ├── fwd_bwd_no_pipelining.py │ │ │ │ ├── fwd_bwd_pipelining_with_interleaving.py │ │ │ │ └── fwd_bwd_pipelining_without_interleaving.py │ │ │ └── utils.py │ │ │ ├── tensor_parallel │ │ │ ├── __init__.py │ │ │ ├── cross_entropy.py │ │ │ ├── data.py │ │ │ ├── layers.py │ │ │ ├── mappings.py │ │ │ ├── memory.py │ │ │ ├── random.py │ │ │ └── utils.py │ │ │ ├── testing │ │ │ ├── __init__.py │ │ │ ├── arguments.py │ │ │ ├── commons.py │ │ │ ├── distributed_test_base.py │ │ │ ├── global_vars.py │ │ │ ├── standalone_bert.py │ │ │ ├── standalone_gpt.py │ │ │ └── standalone_transformer_lm.py │ │ │ └── utils.py │ ├── csrc │ │ ├── amp_C_frontend.cpp │ │ ├── compat.h │ │ ├── flatten_unflatten.cpp │ │ ├── fused_dense.cpp │ │ ├── fused_dense_cuda.cu │ │ ├── layer_norm_cuda.cpp │ │ ├── layer_norm_cuda_kernel.cu │ │ ├── megatron │ │ │ ├── fused_weight_gradient_dense.cpp │ │ │ ├── fused_weight_gradient_dense_16bit_prec_cuda.cu │ │ │ ├── fused_weight_gradient_dense_cuda.cu │ │ │ ├── generic_scaled_masked_softmax.cpp │ │ │ ├── generic_scaled_masked_softmax.h │ │ │ ├── generic_scaled_masked_softmax_cuda.cu │ │ │ ├── scaled_masked_softmax.cpp │ │ │ ├── scaled_masked_softmax.h │ │ │ ├── scaled_masked_softmax_cuda.cu │ │ │ ├── scaled_upper_triang_masked_softmax.cpp │ │ │ ├── scaled_upper_triang_masked_softmax.h │ │ │ └── scaled_upper_triang_masked_softmax_cuda.cu │ │ ├── mlp.cpp │ │ ├── mlp_cuda.cu │ │ ├── multi_tensor_adagrad.cu │ │ ├── multi_tensor_adam.cu │ │ ├── multi_tensor_apply.cuh │ │ ├── multi_tensor_axpby_kernel.cu │ │ ├── multi_tensor_l2norm_kernel.cu │ │ ├── multi_tensor_l2norm_kernel_mp.cu │ │ ├── multi_tensor_l2norm_scale_kernel.cu │ │ ├── multi_tensor_lamb.cu │ │ ├── multi_tensor_lamb_mp.cu │ │ ├── multi_tensor_lamb_stage_1.cu │ │ ├── multi_tensor_lamb_stage_2.cu │ │ ├── multi_tensor_novograd.cu │ │ ├── multi_tensor_scale_kernel.cu │ │ ├── multi_tensor_sgd_kernel.cu │ │ ├── syncbn.cpp │ │ ├── type_shim.h │ │ └── welford.cu │ ├── dist │ │ └── apex-0.1-py3.8.egg │ ├── docs │ │ ├── Makefile │ │ └── source │ │ │ ├── _static │ │ │ ├── css │ │ │ │ └── pytorch_theme.css │ │ │ └── img │ │ │ │ └── nv-pytorch2.png │ │ │ ├── _templates │ │ │ └── layout.html │ │ │ ├── advanced.rst │ │ │ ├── amp.rst │ │ │ ├── conf.py │ │ │ ├── fp16_utils.rst │ │ │ ├── index.rst │ │ │ ├── layernorm.rst │ │ │ ├── optimizers.rst │ │ │ └── parallel.rst │ ├── examples │ │ ├── README.md │ │ ├── dcgan │ │ │ ├── README.md │ │ │ └── main_amp.py │ │ ├── docker │ │ │ ├── Dockerfile │ │ │ └── README.md │ │ ├── imagenet │ │ │ ├── README.md │ │ │ └── main_amp.py │ │ └── simple │ │ │ └── distributed │ │ │ ├── README.md │ │ │ ├── distributed_data_parallel.py │ │ │ └── run.sh │ ├── requirements.txt │ ├── requirements_dev.txt │ ├── setup.py │ └── tests │ │ ├── L0 │ │ ├── run_amp │ │ │ ├── __init__.py │ │ │ ├── test_add_param_group.py │ │ │ ├── test_basic_casts.py │ │ │ ├── test_cache.py │ │ │ ├── test_checkpointing.py │ │ │ ├── test_fused_sgd.py │ │ │ ├── test_larc.py │ │ │ ├── test_multi_tensor_axpby.py │ │ │ ├── test_multi_tensor_l2norm.py │ │ │ ├── test_multi_tensor_scale.py │ │ │ ├── test_multiple_models_optimizers_losses.py │ │ │ ├── test_promotion.py │ │ │ ├── test_rnn.py │ │ │ └── utils.py │ │ ├── run_fp16util │ │ │ ├── __init__.py │ │ │ └── test_fp16util.py │ │ ├── run_fused_layer_norm │ │ │ └── test_fused_layer_norm.py │ │ ├── run_mlp │ │ │ └── test_mlp.py │ │ ├── run_optimizers │ │ │ ├── __init__.py │ │ │ ├── test_fused_novograd.py │ │ │ ├── test_fused_optimizer.py │ │ │ └── test_lamb.py │ │ ├── run_test.py │ │ └── run_transformer │ │ │ ├── __init__.py │ │ │ ├── gpt_scaling_test.py │ │ │ ├── run_bert_minimal_test.py │ │ │ ├── run_dynamic_batchsize_test.py │ │ │ ├── run_gpt_minimal_test.py │ │ │ ├── test_batch_sampler.py │ │ │ ├── test_cross_entropy.py │ │ │ ├── test_data.py │ │ │ ├── test_fused_softmax.py │ │ │ ├── test_layers.py │ │ │ ├── test_mapping.py │ │ │ ├── test_microbatches.py │ │ │ ├── test_p2p_comm.py │ │ │ ├── test_parallel_state.py │ │ │ ├── test_pipeline_parallel_fwd_bwd.py │ │ │ ├── test_random.py │ │ │ ├── test_transformer_module.py │ │ │ └── test_transformer_utils.py │ │ ├── L1 │ │ ├── common │ │ │ ├── compare.py │ │ │ ├── main_amp.py │ │ │ └── run_test.sh │ │ ├── cross_product │ │ │ └── run.sh │ │ ├── cross_product_distributed │ │ │ └── run.sh │ │ └── transformer │ │ │ └── pipeline_parallel_fwd_bwd_ucc_async.py │ │ ├── distributed │ │ ├── DDP │ │ │ ├── ddp_race_condition_test.py │ │ │ └── run_race_test.sh │ │ ├── amp_master_params │ │ │ ├── amp_master_params.py │ │ │ ├── compare.py │ │ │ └── run.sh │ │ └── synced_batchnorm │ │ │ ├── python_single_gpu_unit_test.py │ │ │ ├── single_gpu_unit_test.py │ │ │ ├── test_batchnorm1d.py │ │ │ ├── test_groups.py │ │ │ ├── two_gpu_test_different_batch_size.py │ │ │ ├── two_gpu_unit_test.py │ │ │ └── unit_test.sh │ │ └── docker_extension_builds │ │ └── run.sh ├── cct │ ├── LICENSE │ ├── configs │ │ ├── datasets │ │ │ ├── cifar10.yml │ │ │ ├── cifar100.yml │ │ │ ├── flowers102.yml │ │ │ └── imagenet.yml │ │ ├── finetuned │ │ │ ├── cct_14-7x2_flowers102.yml │ │ │ └── cct_14-7x2_imagenet384.yml │ │ └── pretrained │ │ │ ├── cct_14-7x2_imagenet.yml │ │ │ ├── cct_7-7x2_flowers102.yml │ │ │ ├── cct_cifar100_HTM.yaml │ │ │ ├── cct_cifar100_HVTM.yaml │ │ │ ├── cct_cifar100_VTM.yaml │ │ │ ├── cct_cifar10_HTM.yaml │ │ │ ├── cct_cifar10_HVTM.yaml │ │ │ ├── cct_cifar10_VTM.yaml │ │ │ ├── vit_lite_7-3x1_cifar100_1500epochs.yml │ │ │ └── vit_lite_7-3x1_cifar100_300epochs.yml │ ├── eval.py │ ├── nlp │ │ └── README.md │ ├── src │ │ ├── __init__.py │ │ ├── cct.py │ │ ├── cvt.py │ │ ├── registry.py │ │ ├── text │ │ │ ├── __init__.py │ │ │ ├── cct.py │ │ │ ├── cvt.py │ │ │ ├── transformer.py │ │ │ └── vit.py │ │ ├── utils │ │ │ ├── __init__.py │ │ │ ├── embedder.py │ │ │ ├── helpers.py │ │ │ ├── stochastic_depth.py │ │ │ ├── tokenizer.py │ │ │ ├── transformers.py │ │ │ ├── utils.py │ │ │ └── utils_analysis.py │ │ └── vit.py │ └── train.py ├── run │ ├── eval_cct_7-3x1_cifar100_HTM.sh │ ├── eval_cct_7-3x1_cifar100_HVTM.sh │ ├── eval_cct_7-3x1_cifar100_VTM.sh │ ├── eval_cct_7-3x1_cifar10_HTM.sh │ ├── eval_cct_7-3x1_cifar10_HVTM.sh │ ├── eval_cct_7-3x1_cifar10_VTM.sh │ ├── eval_vit_imagenet_HTM.sh │ ├── eval_vit_imagenet_HVTM.sh │ ├── eval_vit_imagenet_VTM.sh │ ├── train_cct_7-3x1_cifar100_HTM.sh │ ├── train_cct_7-3x1_cifar100_HVTM.sh │ ├── train_cct_7-3x1_cifar100_VTM.sh │ ├── train_cct_7-3x1_cifar10_HTM.sh │ ├── train_cct_7-3x1_cifar10_HVTM.sh │ ├── train_cct_7-3x1_cifar10_VTM.sh │ ├── train_vit_imagenet_HTM.sh │ ├── train_vit_imagenet_HVTM.sh │ └── train_vit_imagenet_VTM.sh └── vit │ ├── .gitignore │ ├── LICENSE │ ├── models │ ├── configs.py │ ├── modeling.py │ └── modeling_resnet.py │ ├── train.py │ └── utils │ ├── data_utils.py │ ├── dist_util.py │ ├── scheduler.py │ └── utils.py └── tokenmixup ├── __init__.py ├── horizontal.py └── vertical.py /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/README.md -------------------------------------------------------------------------------- /assets/HTM.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/assets/HTM.gif -------------------------------------------------------------------------------- /assets/VTM.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/assets/VTM.gif -------------------------------------------------------------------------------- /env.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/env.yaml -------------------------------------------------------------------------------- /experiments/apex_copy/.github/ISSUE_TEMPLATE/bug_report.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/.github/ISSUE_TEMPLATE/bug_report.md -------------------------------------------------------------------------------- /experiments/apex_copy/.gitmodules: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/.gitmodules -------------------------------------------------------------------------------- /experiments/apex_copy/.nojekyll: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/LICENSE -------------------------------------------------------------------------------- /experiments/apex_copy/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex.egg-info/PKG-INFO: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex.egg-info/PKG-INFO -------------------------------------------------------------------------------- /experiments/apex_copy/apex.egg-info/SOURCES.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex.egg-info/SOURCES.txt -------------------------------------------------------------------------------- /experiments/apex_copy/apex.egg-info/dependency_links.txt: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex.egg-info/top_level.txt: -------------------------------------------------------------------------------- 1 | apex 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/RNN/README.md: -------------------------------------------------------------------------------- 1 | Under construction... 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/RNN/RNNBackend.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/RNN/RNNBackend.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/RNN/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/RNN/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/RNN/cells.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/RNN/cells.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/RNN/models.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/RNN/models.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/_autocast_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/_autocast_utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/__version__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/__version__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/_amp_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/_amp_state.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/_initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/_initialize.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/_process_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/_process_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/amp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/amp.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/compat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/compat.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/frontend.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/frontend.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/handle.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/handle.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/lists/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/lists/functional_overrides.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/lists/functional_overrides.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/lists/tensor_overrides.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/lists/tensor_overrides.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/lists/torch_overrides.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/lists/torch_overrides.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/opt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/opt.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/rnn_compat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/rnn_compat.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/scaler.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/amp/wrap.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/amp/wrap.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/bottleneck/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/bottleneck/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/bottleneck/bottleneck.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/bottleneck/bottleneck.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/bottleneck/halo_exchangers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/bottleneck/halo_exchangers.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/bottleneck/test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/bottleneck/test.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/clip_grad/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/clip_grad/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/clip_grad/clip_grad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/clip_grad/clip_grad.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/conv_bias_relu/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/conv_bias_relu/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/conv_bias_relu/conv_bias_relu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/conv_bias_relu/conv_bias_relu.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/bottleneck/bottleneck.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/bottleneck/bottleneck.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/conv_bias_relu/conv_bias_relu.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/conv_bias_relu/conv_bias_relu.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/cudnn_gbn/cudnn_gbn.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/cudnn_gbn/cudnn_gbn.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/cudnn_gbn/norm_sample.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/cudnn_gbn/norm_sample.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/cudnn_gbn/norm_sample.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/cudnn_gbn/norm_sample.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/fmha_api.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/fmha_api.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/gemm.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/gemm.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/gmem_tile.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/gmem_tile.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/kernel_traits.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/kernel_traits.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/mask.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/mask.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/smem_tile.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/smem_tile.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/softmax.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/softmax.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/utils.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha/utils.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_128_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_128_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_256_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_256_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_384_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_384_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_512_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_fp16_512_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_kernel_1xN_reload.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_kernel_1xN_reload.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_kernel_1xN_reload_nl.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_dgrad_kernel_1xN_reload_nl.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_128_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_128_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_256_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_256_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_384_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_384_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_512_64_kernel.sm80.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_fp16_512_64_kernel.sm80.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_kernel_1xN.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_fprop_kernel_1xN.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_kernel.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_kernel.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_noloop_reduce.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_noloop_reduce.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_utils.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/fmha/src/fmha_utils.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/focal_loss/focal_loss_cuda.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/focal_loss/focal_loss_cuda.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/focal_loss/focal_loss_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/focal_loss/focal_loss_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm_add_relu.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm_add_relu.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm_add_relu.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/batch_norm_add_relu.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/cuda_utils.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/cuda_utils.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/interface.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/interface.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/ipc.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/ipc.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/groupbn/nhwc_batch_norm_kernel.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/groupbn/nhwc_batch_norm_kernel.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/index_mul_2d/index_mul_2d_cuda.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/index_mul_2d/index_mul_2d_cuda.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/index_mul_2d/index_mul_2d_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/index_mul_2d/index_mul_2d_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_api.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_api.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_bwd_kernels.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_bwd_kernels.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_bwd_semi_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_bwd_semi_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_fwd_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_fwd_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_fwd_kernels.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_fwd_kernels.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_kernel_traits.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_kernel_traits.h -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_utils.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/layer_norm/ln_utils.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/additive_masked_softmax_dropout_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/additive_masked_softmax_dropout_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/dropout.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/dropout.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/encdec_multihead_attn_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/encdec_multihead_attn_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/encdec_multihead_attn_norm_add_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/encdec_multihead_attn_norm_add_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/layer_norm.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/layer_norm.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/masked_softmax_dropout_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/masked_softmax_dropout_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/multihead_attn_frontend.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/multihead_attn_frontend.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/philox.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/philox.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_bias_additive_mask_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_bias_additive_mask_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_bias_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_bias_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_norm_add_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/self_multihead_attn_norm_add_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/softmax.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/softmax.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/multihead_attn/strided_batched_gemm.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/multihead_attn/strided_batched_gemm.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_p2p.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_p2p.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_p2p_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_p2p_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_p2p_cuda.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_p2p_cuda.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_version.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_version.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_version_check.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/nccl_p2p/nccl_version_check.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/fused_adam_cuda.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/fused_adam_cuda.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/fused_adam_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/fused_adam_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/fused_lamb_cuda.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/fused_lamb_cuda.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/fused_lamb_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/fused_lamb_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_adam.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_adam.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_adam_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_adam_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_lamb.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_lamb.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_lamb_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/optimizers/multi_tensor_distopt_lamb_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/peer_memory/peer_memory.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/peer_memory/peer_memory.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/peer_memory/peer_memory_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/peer_memory/peer_memory_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/peer_memory/peer_memory_cuda.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/peer_memory/peer_memory_cuda.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/transducer/transducer_joint.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/transducer/transducer_joint.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/transducer/transducer_joint_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/transducer/transducer_joint_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/transducer/transducer_loss.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/transducer/transducer_loss.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/transducer/transducer_loss_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/transducer/transducer_loss_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/xentropy/interface.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/xentropy/interface.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/csrc/xentropy/xentropy_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/csrc/xentropy/xentropy_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/cudnn_gbn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/cudnn_gbn/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/cudnn_gbn/batch_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/cudnn_gbn/batch_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/examples/multihead_attn/func_test_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/examples/multihead_attn/func_test_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/examples/multihead_attn/perf_test_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/examples/multihead_attn/perf_test_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/fmha/__init__.py: -------------------------------------------------------------------------------- 1 | from .fmha import FMHAFun 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/fmha/fmha.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/fmha/fmha.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/focal_loss/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/focal_loss/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/focal_loss/focal_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/focal_loss/focal_loss.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/groupbn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/groupbn/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/groupbn/batch_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/groupbn/batch_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/index_mul_2d/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/index_mul_2d/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/index_mul_2d/index_mul_2d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/index_mul_2d/index_mul_2d.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/layer_norm/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/layer_norm/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/layer_norm/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/layer_norm/layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/MHA_bwd.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/MHA_bwd.png -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/MHA_fwd.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/MHA_fwd.png -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/encdec_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/encdec_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/encdec_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/encdec_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/fast_encdec_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/fast_encdec_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/fast_encdec_multihead_attn_norm_add_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/fast_encdec_multihead_attn_norm_add_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/fast_self_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/fast_self_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/fast_self_multihead_attn_norm_add_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/fast_self_multihead_attn_norm_add_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/mask_softmax_dropout_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/mask_softmax_dropout_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/self_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/self_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/multihead_attn/self_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/multihead_attn/self_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/distributed_fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/distributed_fused_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/distributed_fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/distributed_fused_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/fp16_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/fp16_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/fused_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/fused_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/optimizers/fused_sgd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/optimizers/fused_sgd.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/peer_memory/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/peer_memory/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/peer_memory/peer_halo_exchanger_1d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/peer_memory/peer_halo_exchanger_1d.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/peer_memory/peer_memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/peer_memory/peer_memory.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/COPYRIGHT: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/COPYRIGHT -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/asp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/asp.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_lib.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_lib.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/CUDA_kernels/permutation_search_kernels.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/CUDA_kernels/permutation_search_kernels.cu -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/call_permutation_search_kernels.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/call_permutation_search_kernels.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/channel_swap.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/channel_swap.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/exhaustive_search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/exhaustive_search.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/permutation_utilities.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_search_kernels/permutation_utilities.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_tests/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_tests/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_tests/ablation_studies.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_tests/ablation_studies.sh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_tests/permutation_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_tests/permutation_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_tests/runtime_table.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_tests/runtime_table.sh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/permutation_tests/unstructured_study.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/permutation_tests/unstructured_study.sh -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/sparse_masklib.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/sparse_masklib.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/test/checkpointing_test_part1.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/test/checkpointing_test_part1.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/test/checkpointing_test_part2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/test/checkpointing_test_part2.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/test/checkpointing_test_reference.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/test/checkpointing_test_reference.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/sparsity/test/toy_problem.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/sparsity/test/toy_problem.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/bottleneck/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/bottleneck/test_bottleneck_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/bottleneck/test_bottleneck_module.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/clip_grad/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/clip_grad/test_clip_grad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/clip_grad/test_clip_grad.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/conv_bias_relu/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/conv_bias_relu/test_conv_bias_relu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/conv_bias_relu/test_conv_bias_relu.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/cudnn_gbn/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/cudnn_gbn/test_cudnn_gbn_with_two_gpus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/cudnn_gbn/test_cudnn_gbn_with_two_gpus.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/fmha/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/fmha/test_fmha.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/fmha/test_fmha.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/focal_loss/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/focal_loss/test_focal_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/focal_loss/test_focal_loss.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/fused_dense/test_fused_dense.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/fused_dense/test_fused_dense.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/index_mul_2d/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/index_mul_2d/test_index_mul_2d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/index_mul_2d/test_index_mul_2d.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/layer_norm/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/layer_norm/test_fast_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/layer_norm/test_fast_layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/test_encdec_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/multihead_attn/test_encdec_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/test_encdec_multihead_attn_norm_add.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/multihead_attn/test_encdec_multihead_attn_norm_add.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/test_fast_self_multihead_attn_bias.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/multihead_attn/test_fast_self_multihead_attn_bias.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/test_mha_fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/multihead_attn/test_mha_fused_softmax.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/test_self_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/multihead_attn/test_self_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/multihead_attn/test_self_multihead_attn_norm_add.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/multihead_attn/test_self_multihead_attn_norm_add.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/optimizers/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/optimizers/test_dist_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/optimizers/test_dist_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/peer_memory/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/peer_memory/test_peer_halo_exchange_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/peer_memory/test_peer_halo_exchange_module.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/transducer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/transducer/test_transducer_joint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/transducer/test_transducer_joint.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/transducer/test_transducer_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/transducer/test_transducer_loss.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/xentropy/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/test/xentropy/test_label_smoothing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/test/xentropy/test_label_smoothing.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/transducer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/transducer/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/transducer/_transducer_ref.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/transducer/_transducer_ref.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/transducer/transducer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/transducer/transducer.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/xentropy/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/xentropy/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/contrib/xentropy/softmax_xentropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/contrib/xentropy/softmax_xentropy.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fp16_utils/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/fp16_utils/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fp16_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/fp16_utils/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fp16_utils/fp16_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/fp16_utils/fp16_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fp16_utils/fp16util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/fp16_utils/fp16util.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fp16_utils/loss_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/fp16_utils/loss_scaler.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fused_dense/__init__.py: -------------------------------------------------------------------------------- 1 | from .fused_dense import * 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/fused_dense/fused_dense.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/fused_dense/fused_dense.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/mlp/__init__.py: -------------------------------------------------------------------------------- 1 | from .mlp import * 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/mlp/mlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/mlp/mlp.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/multi_tensor_apply/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/multi_tensor_apply/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/multi_tensor_apply/multi_tensor_apply.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/multi_tensor_apply/multi_tensor_apply.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/normalization/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/normalization/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/normalization/fused_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/normalization/fused_layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/fused_adagrad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/fused_adagrad.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/fused_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/fused_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/fused_mixed_precision_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/fused_mixed_precision_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/fused_novograd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/fused_novograd.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/optimizers/fused_sgd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/optimizers/fused_sgd.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/LARC.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/LARC.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/distributed.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/multiproc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/multiproc.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/optimized_sync_batchnorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/optimized_sync_batchnorm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/optimized_sync_batchnorm_kernel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/optimized_sync_batchnorm_kernel.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/sync_batchnorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/sync_batchnorm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/parallel/sync_batchnorm_kernel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/parallel/sync_batchnorm_kernel.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/_data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/_data/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/_data/_batchsampler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/_data/_batchsampler.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/amp/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/amp/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/amp/grad_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/amp/grad_scaler.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/enums.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/functional/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/functional/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/functional/fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/functional/fused_softmax.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/layers/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/layers/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/layers/layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/log_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/log_util.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/microbatches.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/microbatches.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/parallel_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/parallel_state.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/_timers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/_timers.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/p2p_communication.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/p2p_communication.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/common.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/fwd_bwd_no_pipelining.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/fwd_bwd_no_pipelining.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_with_interleaving.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_with_interleaving.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_without_interleaving.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_without_interleaving.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/pipeline_parallel/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/pipeline_parallel/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/cross_entropy.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/data.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/layers.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/mappings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/mappings.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/memory.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/random.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/tensor_parallel/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/tensor_parallel/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/arguments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/arguments.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/commons.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/commons.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/distributed_test_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/distributed_test_base.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/global_vars.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/global_vars.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/standalone_bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/standalone_bert.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/standalone_gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/standalone_gpt.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/testing/standalone_transformer_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/testing/standalone_transformer_lm.py -------------------------------------------------------------------------------- /experiments/apex_copy/apex/transformer/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/apex/transformer/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/RNN/RNNBackend.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/RNN/RNNBackend.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/RNN/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/RNN/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/RNN/cells.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/RNN/cells.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/RNN/models.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/RNN/models.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/_autocast_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/_autocast_utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/__version__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/__version__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/_amp_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/_amp_state.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/_initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/_initialize.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/_process_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/_process_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/amp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/amp.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/compat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/compat.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/frontend.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/frontend.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/handle.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/handle.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/lists/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/lists/functional_overrides.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/lists/functional_overrides.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/lists/tensor_overrides.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/lists/tensor_overrides.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/lists/torch_overrides.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/lists/torch_overrides.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/opt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/opt.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/rnn_compat.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/rnn_compat.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/scaler.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/amp/wrap.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/amp/wrap.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/bottleneck/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/bottleneck/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/bottleneck/bottleneck.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/bottleneck/bottleneck.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/bottleneck/halo_exchangers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/bottleneck/halo_exchangers.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/bottleneck/test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/bottleneck/test.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/clip_grad/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/clip_grad/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/clip_grad/clip_grad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/clip_grad/clip_grad.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/conv_bias_relu/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/conv_bias_relu/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/conv_bias_relu/conv_bias_relu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/conv_bias_relu/conv_bias_relu.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/cudnn_gbn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/cudnn_gbn/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/cudnn_gbn/batch_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/cudnn_gbn/batch_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/fmha/__init__.py: -------------------------------------------------------------------------------- 1 | from .fmha import FMHAFun 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/fmha/fmha.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/fmha/fmha.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/focal_loss/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/focal_loss/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/focal_loss/focal_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/focal_loss/focal_loss.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/groupbn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/groupbn/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/groupbn/batch_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/groupbn/batch_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/index_mul_2d/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/index_mul_2d/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/index_mul_2d/index_mul_2d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/index_mul_2d/index_mul_2d.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/layer_norm/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/layer_norm/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/layer_norm/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/layer_norm/layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/encdec_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/encdec_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/encdec_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/encdec_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_encdec_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_encdec_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_encdec_multihead_attn_norm_add_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_encdec_multihead_attn_norm_add_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_self_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_self_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_self_multihead_attn_norm_add_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/fast_self_multihead_attn_norm_add_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/mask_softmax_dropout_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/mask_softmax_dropout_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/self_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/self_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/multihead_attn/self_multihead_attn_func.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/multihead_attn/self_multihead_attn_func.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/distributed_fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/distributed_fused_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/distributed_fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/distributed_fused_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/fp16_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/fp16_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/fused_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/fused_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/optimizers/fused_sgd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/optimizers/fused_sgd.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/peer_memory/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/peer_memory/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/peer_memory/peer_halo_exchanger_1d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/peer_memory/peer_halo_exchanger_1d.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/peer_memory/peer_memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/peer_memory/peer_memory.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/asp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/asp.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_lib.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_lib.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/call_permutation_search_kernels.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/call_permutation_search_kernels.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/channel_swap.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/channel_swap.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/exhaustive_search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/exhaustive_search.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/permutation_utilities.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/permutation_search_kernels/permutation_utilities.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/sparsity/sparse_masklib.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/sparsity/sparse_masklib.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/bottleneck/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/bottleneck/test_bottleneck_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/bottleneck/test_bottleneck_module.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/clip_grad/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/clip_grad/test_clip_grad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/clip_grad/test_clip_grad.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/conv_bias_relu/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/conv_bias_relu/test_conv_bias_relu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/conv_bias_relu/test_conv_bias_relu.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/cudnn_gbn/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/cudnn_gbn/test_cudnn_gbn_with_two_gpus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/cudnn_gbn/test_cudnn_gbn_with_two_gpus.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/fmha/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/fmha/test_fmha.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/fmha/test_fmha.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/focal_loss/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/focal_loss/test_focal_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/focal_loss/test_focal_loss.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/index_mul_2d/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/index_mul_2d/test_index_mul_2d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/index_mul_2d/test_index_mul_2d.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/layer_norm/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/layer_norm/test_fast_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/layer_norm/test_fast_layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_encdec_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_encdec_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_encdec_multihead_attn_norm_add.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_encdec_multihead_attn_norm_add.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_fast_self_multihead_attn_bias.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_fast_self_multihead_attn_bias.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_mha_fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_mha_fused_softmax.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_self_multihead_attn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_self_multihead_attn.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_self_multihead_attn_norm_add.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/multihead_attn/test_self_multihead_attn_norm_add.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/optimizers/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/optimizers/test_dist_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/optimizers/test_dist_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/peer_memory/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/peer_memory/test_peer_halo_exchange_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/peer_memory/test_peer_halo_exchange_module.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/transducer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/transducer/test_transducer_joint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/transducer/test_transducer_joint.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/transducer/test_transducer_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/transducer/test_transducer_loss.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/xentropy/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/test/xentropy/test_label_smoothing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/test/xentropy/test_label_smoothing.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/transducer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/transducer/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/transducer/_transducer_ref.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/transducer/_transducer_ref.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/transducer/transducer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/transducer/transducer.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/xentropy/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/xentropy/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/contrib/xentropy/softmax_xentropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/contrib/xentropy/softmax_xentropy.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/fp16_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/fp16_utils/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/fp16_utils/fp16_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/fp16_utils/fp16_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/fp16_utils/fp16util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/fp16_utils/fp16util.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/fp16_utils/loss_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/fp16_utils/loss_scaler.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/fused_dense/__init__.py: -------------------------------------------------------------------------------- 1 | from .fused_dense import * 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/fused_dense/fused_dense.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/fused_dense/fused_dense.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/mlp/__init__.py: -------------------------------------------------------------------------------- 1 | from .mlp import * 2 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/mlp/mlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/mlp/mlp.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/multi_tensor_apply/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/multi_tensor_apply/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/multi_tensor_apply/multi_tensor_apply.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/multi_tensor_apply/multi_tensor_apply.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/normalization/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/normalization/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/normalization/fused_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/normalization/fused_layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/fused_adagrad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/fused_adagrad.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/fused_adam.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/fused_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/fused_mixed_precision_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/fused_mixed_precision_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/fused_novograd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/fused_novograd.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/optimizers/fused_sgd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/optimizers/fused_sgd.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/LARC.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/LARC.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/distributed.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/multiproc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/multiproc.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/optimized_sync_batchnorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/optimized_sync_batchnorm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/optimized_sync_batchnorm_kernel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/optimized_sync_batchnorm_kernel.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/sync_batchnorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/sync_batchnorm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/parallel/sync_batchnorm_kernel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/parallel/sync_batchnorm_kernel.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/_data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/_data/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/_data/_batchsampler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/_data/_batchsampler.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/amp/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/amp/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/amp/grad_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/amp/grad_scaler.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/enums.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/functional/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/functional/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/functional/fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/functional/fused_softmax.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/layers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/layers/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/layers/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/layers/layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/log_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/log_util.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/microbatches.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/microbatches.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/parallel_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/parallel_state.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/_timers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/_timers.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/p2p_communication.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/p2p_communication.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/common.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/fwd_bwd_no_pipelining.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/fwd_bwd_no_pipelining.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_with_interleaving.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_with_interleaving.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_without_interleaving.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_without_interleaving.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/pipeline_parallel/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/__init__.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/cross_entropy.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/data.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/layers.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/mappings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/mappings.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/memory.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/random.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/tensor_parallel/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/arguments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/arguments.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/commons.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/commons.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/distributed_test_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/distributed_test_base.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/global_vars.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/global_vars.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/standalone_bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/standalone_bert.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/standalone_gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/standalone_gpt.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/testing/standalone_transformer_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/testing/standalone_transformer_lm.py -------------------------------------------------------------------------------- /experiments/apex_copy/build/lib/apex/transformer/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/build/lib/apex/transformer/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/amp_C_frontend.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/amp_C_frontend.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/compat.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/compat.h -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/flatten_unflatten.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/flatten_unflatten.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/fused_dense.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/fused_dense.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/fused_dense_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/fused_dense_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/layer_norm_cuda.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/layer_norm_cuda.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/layer_norm_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/layer_norm_cuda_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/fused_weight_gradient_dense.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/fused_weight_gradient_dense.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/fused_weight_gradient_dense_16bit_prec_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/fused_weight_gradient_dense_16bit_prec_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/fused_weight_gradient_dense_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/fused_weight_gradient_dense_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/generic_scaled_masked_softmax.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/generic_scaled_masked_softmax.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/generic_scaled_masked_softmax.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/generic_scaled_masked_softmax.h -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/generic_scaled_masked_softmax_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/generic_scaled_masked_softmax_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/scaled_masked_softmax.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/scaled_masked_softmax.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/scaled_masked_softmax.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/scaled_masked_softmax.h -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/scaled_masked_softmax_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/scaled_masked_softmax_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/scaled_upper_triang_masked_softmax.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/scaled_upper_triang_masked_softmax.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/scaled_upper_triang_masked_softmax.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/scaled_upper_triang_masked_softmax.h -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/megatron/scaled_upper_triang_masked_softmax_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/megatron/scaled_upper_triang_masked_softmax_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/mlp.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/mlp.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/mlp_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/mlp_cuda.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_adagrad.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_adagrad.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_adam.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_adam.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_apply.cuh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_apply.cuh -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_axpby_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_axpby_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_l2norm_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_l2norm_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_l2norm_kernel_mp.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_l2norm_kernel_mp.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_l2norm_scale_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_l2norm_scale_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_lamb.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_lamb.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_lamb_mp.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_lamb_mp.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_lamb_stage_1.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_lamb_stage_1.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_lamb_stage_2.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_lamb_stage_2.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_novograd.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_novograd.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_scale_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_scale_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/multi_tensor_sgd_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/multi_tensor_sgd_kernel.cu -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/syncbn.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/syncbn.cpp -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/type_shim.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/type_shim.h -------------------------------------------------------------------------------- /experiments/apex_copy/csrc/welford.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/csrc/welford.cu -------------------------------------------------------------------------------- /experiments/apex_copy/dist/apex-0.1-py3.8.egg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/dist/apex-0.1-py3.8.egg -------------------------------------------------------------------------------- /experiments/apex_copy/docs/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/Makefile -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/_static/css/pytorch_theme.css: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/_static/css/pytorch_theme.css -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/_static/img/nv-pytorch2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/_static/img/nv-pytorch2.png -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/_templates/layout.html: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/_templates/layout.html -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/advanced.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/advanced.rst -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/amp.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/amp.rst -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/conf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/conf.py -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/fp16_utils.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/fp16_utils.rst -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/index.rst -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/layernorm.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/layernorm.rst -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/optimizers.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/optimizers.rst -------------------------------------------------------------------------------- /experiments/apex_copy/docs/source/parallel.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/docs/source/parallel.rst -------------------------------------------------------------------------------- /experiments/apex_copy/examples/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/examples/dcgan/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/dcgan/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/examples/dcgan/main_amp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/dcgan/main_amp.py -------------------------------------------------------------------------------- /experiments/apex_copy/examples/docker/Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/docker/Dockerfile -------------------------------------------------------------------------------- /experiments/apex_copy/examples/docker/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/docker/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/examples/imagenet/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/imagenet/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/examples/imagenet/main_amp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/imagenet/main_amp.py -------------------------------------------------------------------------------- /experiments/apex_copy/examples/simple/distributed/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/simple/distributed/README.md -------------------------------------------------------------------------------- /experiments/apex_copy/examples/simple/distributed/distributed_data_parallel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/simple/distributed/distributed_data_parallel.py -------------------------------------------------------------------------------- /experiments/apex_copy/examples/simple/distributed/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/examples/simple/distributed/run.sh -------------------------------------------------------------------------------- /experiments/apex_copy/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/requirements.txt -------------------------------------------------------------------------------- /experiments/apex_copy/requirements_dev.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/requirements_dev.txt -------------------------------------------------------------------------------- /experiments/apex_copy/setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/setup.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_add_param_group.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_add_param_group.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_basic_casts.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_basic_casts.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_cache.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_cache.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_checkpointing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_checkpointing.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_fused_sgd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_fused_sgd.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_larc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_larc.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_multi_tensor_axpby.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_multi_tensor_axpby.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_multi_tensor_l2norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_multi_tensor_l2norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_multi_tensor_scale.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_multi_tensor_scale.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_multiple_models_optimizers_losses.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_multiple_models_optimizers_losses.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_promotion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_promotion.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/test_rnn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/test_rnn.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_amp/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_amp/utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_fp16util/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_fp16util/test_fp16util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_fp16util/test_fp16util.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_fused_layer_norm/test_fused_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_fused_layer_norm/test_fused_layer_norm.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_mlp/test_mlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_mlp/test_mlp.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_optimizers/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_optimizers/test_fused_novograd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_optimizers/test_fused_novograd.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_optimizers/test_fused_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_optimizers/test_fused_optimizer.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_optimizers/test_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_optimizers/test_lamb.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/gpt_scaling_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/gpt_scaling_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/run_bert_minimal_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/run_bert_minimal_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/run_dynamic_batchsize_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/run_dynamic_batchsize_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/run_gpt_minimal_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/run_gpt_minimal_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_batch_sampler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_batch_sampler.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_cross_entropy.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_data.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_fused_softmax.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_layers.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_mapping.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_mapping.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_microbatches.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_microbatches.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_p2p_comm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_p2p_comm.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_parallel_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_parallel_state.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_pipeline_parallel_fwd_bwd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_pipeline_parallel_fwd_bwd.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_random.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_transformer_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_transformer_module.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L0/run_transformer/test_transformer_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L0/run_transformer/test_transformer_utils.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L1/common/compare.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L1/common/compare.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L1/common/main_amp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L1/common/main_amp.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L1/common/run_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L1/common/run_test.sh -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L1/cross_product/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L1/cross_product/run.sh -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L1/cross_product_distributed/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L1/cross_product_distributed/run.sh -------------------------------------------------------------------------------- /experiments/apex_copy/tests/L1/transformer/pipeline_parallel_fwd_bwd_ucc_async.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/L1/transformer/pipeline_parallel_fwd_bwd_ucc_async.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/DDP/ddp_race_condition_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/DDP/ddp_race_condition_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/DDP/run_race_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/DDP/run_race_test.sh -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/amp_master_params/amp_master_params.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/amp_master_params/amp_master_params.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/amp_master_params/compare.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/amp_master_params/compare.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/amp_master_params/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/amp_master_params/run.sh -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/python_single_gpu_unit_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/python_single_gpu_unit_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/single_gpu_unit_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/single_gpu_unit_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/test_batchnorm1d.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/test_batchnorm1d.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/test_groups.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/test_groups.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/two_gpu_test_different_batch_size.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/two_gpu_test_different_batch_size.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/two_gpu_unit_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/two_gpu_unit_test.py -------------------------------------------------------------------------------- /experiments/apex_copy/tests/distributed/synced_batchnorm/unit_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/distributed/synced_batchnorm/unit_test.sh -------------------------------------------------------------------------------- /experiments/apex_copy/tests/docker_extension_builds/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/apex_copy/tests/docker_extension_builds/run.sh -------------------------------------------------------------------------------- /experiments/cct/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/LICENSE -------------------------------------------------------------------------------- /experiments/cct/configs/datasets/cifar10.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/datasets/cifar10.yml -------------------------------------------------------------------------------- /experiments/cct/configs/datasets/cifar100.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/datasets/cifar100.yml -------------------------------------------------------------------------------- /experiments/cct/configs/datasets/flowers102.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/datasets/flowers102.yml -------------------------------------------------------------------------------- /experiments/cct/configs/datasets/imagenet.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/datasets/imagenet.yml -------------------------------------------------------------------------------- /experiments/cct/configs/finetuned/cct_14-7x2_flowers102.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/finetuned/cct_14-7x2_flowers102.yml -------------------------------------------------------------------------------- /experiments/cct/configs/finetuned/cct_14-7x2_imagenet384.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/finetuned/cct_14-7x2_imagenet384.yml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_14-7x2_imagenet.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_14-7x2_imagenet.yml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_7-7x2_flowers102.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_7-7x2_flowers102.yml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_cifar100_HTM.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_cifar100_HTM.yaml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_cifar100_HVTM.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_cifar100_HVTM.yaml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_cifar100_VTM.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_cifar100_VTM.yaml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_cifar10_HTM.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_cifar10_HTM.yaml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_cifar10_HVTM.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_cifar10_HVTM.yaml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/cct_cifar10_VTM.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/cct_cifar10_VTM.yaml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/vit_lite_7-3x1_cifar100_1500epochs.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/vit_lite_7-3x1_cifar100_1500epochs.yml -------------------------------------------------------------------------------- /experiments/cct/configs/pretrained/vit_lite_7-3x1_cifar100_300epochs.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/configs/pretrained/vit_lite_7-3x1_cifar100_300epochs.yml -------------------------------------------------------------------------------- /experiments/cct/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/eval.py -------------------------------------------------------------------------------- /experiments/cct/nlp/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/nlp/README.md -------------------------------------------------------------------------------- /experiments/cct/src/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/__init__.py -------------------------------------------------------------------------------- /experiments/cct/src/cct.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/cct.py -------------------------------------------------------------------------------- /experiments/cct/src/cvt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/cvt.py -------------------------------------------------------------------------------- /experiments/cct/src/registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/registry.py -------------------------------------------------------------------------------- /experiments/cct/src/text/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/text/__init__.py -------------------------------------------------------------------------------- /experiments/cct/src/text/cct.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/text/cct.py -------------------------------------------------------------------------------- /experiments/cct/src/text/cvt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/text/cvt.py -------------------------------------------------------------------------------- /experiments/cct/src/text/transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/text/transformer.py -------------------------------------------------------------------------------- /experiments/cct/src/text/vit.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/text/vit.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/cct/src/utils/embedder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/embedder.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/helpers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/helpers.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/stochastic_depth.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/stochastic_depth.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/tokenizer.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/transformers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/transformers.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/utils.py -------------------------------------------------------------------------------- /experiments/cct/src/utils/utils_analysis.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/utils/utils_analysis.py -------------------------------------------------------------------------------- /experiments/cct/src/vit.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/src/vit.py -------------------------------------------------------------------------------- /experiments/cct/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/cct/train.py -------------------------------------------------------------------------------- /experiments/run/eval_cct_7-3x1_cifar100_HTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_cct_7-3x1_cifar100_HTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_cct_7-3x1_cifar100_HVTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_cct_7-3x1_cifar100_HVTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_cct_7-3x1_cifar100_VTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_cct_7-3x1_cifar100_VTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_cct_7-3x1_cifar10_HTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_cct_7-3x1_cifar10_HTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_cct_7-3x1_cifar10_HVTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_cct_7-3x1_cifar10_HVTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_cct_7-3x1_cifar10_VTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_cct_7-3x1_cifar10_VTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_vit_imagenet_HTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_vit_imagenet_HTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_vit_imagenet_HVTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_vit_imagenet_HVTM.sh -------------------------------------------------------------------------------- /experiments/run/eval_vit_imagenet_VTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/eval_vit_imagenet_VTM.sh -------------------------------------------------------------------------------- /experiments/run/train_cct_7-3x1_cifar100_HTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_cct_7-3x1_cifar100_HTM.sh -------------------------------------------------------------------------------- /experiments/run/train_cct_7-3x1_cifar100_HVTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_cct_7-3x1_cifar100_HVTM.sh -------------------------------------------------------------------------------- /experiments/run/train_cct_7-3x1_cifar100_VTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_cct_7-3x1_cifar100_VTM.sh -------------------------------------------------------------------------------- /experiments/run/train_cct_7-3x1_cifar10_HTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_cct_7-3x1_cifar10_HTM.sh -------------------------------------------------------------------------------- /experiments/run/train_cct_7-3x1_cifar10_HVTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_cct_7-3x1_cifar10_HVTM.sh -------------------------------------------------------------------------------- /experiments/run/train_cct_7-3x1_cifar10_VTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_cct_7-3x1_cifar10_VTM.sh -------------------------------------------------------------------------------- /experiments/run/train_vit_imagenet_HTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_vit_imagenet_HTM.sh -------------------------------------------------------------------------------- /experiments/run/train_vit_imagenet_HVTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_vit_imagenet_HVTM.sh -------------------------------------------------------------------------------- /experiments/run/train_vit_imagenet_VTM.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/run/train_vit_imagenet_VTM.sh -------------------------------------------------------------------------------- /experiments/vit/.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/.gitignore -------------------------------------------------------------------------------- /experiments/vit/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/LICENSE -------------------------------------------------------------------------------- /experiments/vit/models/configs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/models/configs.py -------------------------------------------------------------------------------- /experiments/vit/models/modeling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/models/modeling.py -------------------------------------------------------------------------------- /experiments/vit/models/modeling_resnet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/models/modeling_resnet.py -------------------------------------------------------------------------------- /experiments/vit/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/train.py -------------------------------------------------------------------------------- /experiments/vit/utils/data_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/utils/data_utils.py -------------------------------------------------------------------------------- /experiments/vit/utils/dist_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/utils/dist_util.py -------------------------------------------------------------------------------- /experiments/vit/utils/scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/utils/scheduler.py -------------------------------------------------------------------------------- /experiments/vit/utils/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/experiments/vit/utils/utils.py -------------------------------------------------------------------------------- /tokenmixup/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/tokenmixup/__init__.py -------------------------------------------------------------------------------- /tokenmixup/horizontal.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/tokenmixup/horizontal.py -------------------------------------------------------------------------------- /tokenmixup/vertical.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/mlvlab/TokenMixup/HEAD/tokenmixup/vertical.py --------------------------------------------------------------------------------