LICENSE
README.md
setup.py
apex/__init__.py
apex/_autocast_utils.py
apex.egg-info/PKG-INFO
apex.egg-info/SOURCES.txt
apex.egg-info/dependency_links.txt
apex.egg-info/top_level.txt
apex/RNN/RNNBackend.py
apex/RNN/__init__.py
apex/RNN/cells.py
apex/RNN/models.py
apex/amp/__init__.py
apex/amp/__version__.py
apex/amp/_amp_state.py
apex/amp/_initialize.py
apex/amp/_process_optimizer.py
apex/amp/amp.py
apex/amp/compat.py
apex/amp/frontend.py
apex/amp/handle.py
apex/amp/opt.py
apex/amp/rnn_compat.py
apex/amp/scaler.py
apex/amp/utils.py
apex/amp/wrap.py
apex/amp/lists/__init__.py
apex/amp/lists/functional_overrides.py
apex/amp/lists/tensor_overrides.py
apex/amp/lists/torch_overrides.py
apex/contrib/__init__.py
apex/contrib/bottleneck/__init__.py
apex/contrib/bottleneck/bottleneck.py
apex/contrib/bottleneck/bottleneck_module_test.py
apex/contrib/bottleneck/test.py
apex/contrib/fmha/__init__.py
apex/contrib/fmha/fmha.py
apex/contrib/groupbn/__init__.py
apex/contrib/groupbn/batch_norm.py
apex/contrib/layer_norm/__init__.py
apex/contrib/layer_norm/layer_norm.py
apex/contrib/multihead_attn/__init__.py
apex/contrib/multihead_attn/encdec_multihead_attn.py
apex/contrib/multihead_attn/encdec_multihead_attn_func.py
apex/contrib/multihead_attn/fast_encdec_multihead_attn_func.py
apex/contrib/multihead_attn/fast_encdec_multihead_attn_norm_add_func.py
apex/contrib/multihead_attn/fast_self_multihead_attn_func.py
apex/contrib/multihead_attn/fast_self_multihead_attn_norm_add_func.py
apex/contrib/multihead_attn/mask_softmax_dropout_func.py
apex/contrib/multihead_attn/self_multihead_attn.py
apex/contrib/multihead_attn/self_multihead_attn_func.py
apex/contrib/optimizers/__init__.py
apex/contrib/optimizers/distributed_fused_adam.py
apex/contrib/optimizers/distributed_fused_adam_v2.py
apex/contrib/optimizers/distributed_fused_adam_v3.py
apex/contrib/optimizers/distributed_fused_lamb.py
apex/contrib/optimizers/fp16_optimizer.py
apex/contrib/optimizers/fused_adam.py
apex/contrib/optimizers/fused_lamb.py
apex/contrib/optimizers/fused_sgd.py
apex/contrib/sparsity/__init__.py
apex/contrib/sparsity/asp.py
apex/contrib/sparsity/permutation_lib.py
apex/contrib/sparsity/sparse_masklib.py
apex/contrib/sparsity/permutation_search_kernels/__init__.py
apex/contrib/sparsity/permutation_search_kernels/call_permutation_search_kernels.py
apex/contrib/sparsity/permutation_search_kernels/exhaustive_search.py
apex/contrib/sparsity/permutation_search_kernels/permutation_utilities.py
apex/contrib/transducer/__init__.py
apex/contrib/transducer/transducer.py
apex/contrib/xentropy/__init__.py
apex/contrib/xentropy/softmax_xentropy.py
apex/fp16_utils/__init__.py
apex/fp16_utils/fp16_optimizer.py
apex/fp16_utils/fp16util.py
apex/fp16_utils/loss_scaler.py
apex/fused_dense/__init__.py
apex/fused_dense/fused_dense.py
apex/mlp/__init__.py
apex/mlp/mlp.py
apex/multi_tensor_apply/__init__.py
apex/multi_tensor_apply/multi_tensor_apply.py
apex/normalization/__init__.py
apex/normalization/fused_layer_norm.py
apex/optimizers/__init__.py
apex/optimizers/fused_adagrad.py
apex/optimizers/fused_adam.py
apex/optimizers/fused_lamb.py
apex/optimizers/fused_mixed_precision_lamb.py
apex/optimizers/fused_novograd.py
apex/optimizers/fused_sgd.py
apex/parallel/LARC.py
apex/parallel/__init__.py
apex/parallel/distributed.py
apex/parallel/multiproc.py
apex/parallel/optimized_sync_batchnorm.py
apex/parallel/optimized_sync_batchnorm_kernel.py
apex/parallel/sync_batchnorm.py
apex/parallel/sync_batchnorm_kernel.py
apex/pyprof/__init__.py
apex/pyprof/nvtx/__init__.py
apex/pyprof/nvtx/nvmarker.py
apex/pyprof/parse/__init__.py
apex/pyprof/parse/__main__.py
apex/pyprof/parse/db.py
apex/pyprof/parse/kernel.py
apex/pyprof/parse/nvvp.py
apex/pyprof/parse/parse.py
apex/pyprof/prof/__init__.py
apex/pyprof/prof/__main__.py
apex/pyprof/prof/activation.py
apex/pyprof/prof/base.py
apex/pyprof/prof/blas.py
apex/pyprof/prof/conv.py
apex/pyprof/prof/convert.py
apex/pyprof/prof/data.py
apex/pyprof/prof/dropout.py
apex/pyprof/prof/embedding.py
apex/pyprof/prof/index_slice_join_mutate.py
apex/pyprof/prof/linear.py
apex/pyprof/prof/loss.py
apex/pyprof/prof/misc.py
apex/pyprof/prof/normalization.py
apex/pyprof/prof/optim.py
apex/pyprof/prof/output.py
apex/pyprof/prof/pointwise.py
apex/pyprof/prof/pooling.py
apex/pyprof/prof/prof.py
apex/pyprof/prof/randomSample.py
apex/pyprof/prof/recurrentCell.py
apex/pyprof/prof/reduction.py
apex/pyprof/prof/softmax.py
apex/pyprof/prof/usage.py
apex/pyprof/prof/utility.py
apex/reparameterization/__init__.py
apex/reparameterization/reparameterization.py
apex/reparameterization/weight_norm.py
apex/transformer/__init__.py
apex/transformer/enums.py
apex/transformer/log_util.py
apex/transformer/microbatches.py
apex/transformer/parallel_state.py
apex/transformer/utils.py
apex/transformer/_data/__init__.py
apex/transformer/_data/_batchsampler.py
apex/transformer/amp/__init__.py
apex/transformer/amp/grad_scaler.py
apex/transformer/functional/__init__.py
apex/transformer/functional/fused_softmax.py
apex/transformer/pipeline_parallel/__init__.py
apex/transformer/pipeline_parallel/_timers.py
apex/transformer/pipeline_parallel/p2p_communication.py
apex/transformer/pipeline_parallel/utils.py
apex/transformer/pipeline_parallel/schedules/__init__.py
apex/transformer/pipeline_parallel/schedules/common.py
apex/transformer/pipeline_parallel/schedules/fwd_bwd_no_pipelining.py
apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_with_interleaving.py
apex/transformer/pipeline_parallel/schedules/fwd_bwd_pipelining_without_interleaving.py
apex/transformer/tensor_parallel/__init__.py
apex/transformer/tensor_parallel/cross_entropy.py
apex/transformer/tensor_parallel/data.py
apex/transformer/tensor_parallel/layers.py
apex/transformer/tensor_parallel/mappings.py
apex/transformer/tensor_parallel/memory.py
apex/transformer/tensor_parallel/random.py
apex/transformer/tensor_parallel/utils.py
apex/transformer/testing/__init__.py
apex/transformer/testing/arguments.py
apex/transformer/testing/commons.py
apex/transformer/testing/global_vars.py
apex/transformer/testing/standalone_bert.py
apex/transformer/testing/standalone_gpt.py
csrc/amp_C_frontend.cpp
csrc/flatten_unflatten.cpp
csrc/fused_dense.cpp
csrc/fused_dense_cuda.cu
csrc/layer_norm_cuda.cpp
csrc/layer_norm_cuda_kernel.cu
csrc/mlp.cpp
csrc/mlp_cuda.cu
csrc/multi_tensor_adagrad.cu
csrc/multi_tensor_adam.cu
csrc/multi_tensor_axpby_kernel.cu
csrc/multi_tensor_l2norm_kernel.cu
csrc/multi_tensor_l2norm_kernel_mp.cu
csrc/multi_tensor_l2norm_scale_kernel.cu
csrc/multi_tensor_lamb.cu
csrc/multi_tensor_lamb_mp.cu
csrc/multi_tensor_lamb_stage_1.cu
csrc/multi_tensor_lamb_stage_2.cu
csrc/multi_tensor_novograd.cu
csrc/multi_tensor_scale_kernel.cu
csrc/multi_tensor_sgd_kernel.cu
csrc/syncbn.cpp
csrc/welford.cu
csrc/megatron/scaled_masked_softmax.cpp
csrc/megatron/scaled_masked_softmax_cuda.cu
csrc/megatron/scaled_upper_triang_masked_softmax.cpp
csrc/megatron/scaled_upper_triang_masked_softmax_cuda.cu