LICENSE
MANIFEST.in
README.md
build.txt
setup.cfg
setup.py
version.txt
bin/deepspeed
bin/deepspeed.pt
bin/ds
bin/ds_elastic
bin/ds_report
bin/ds_ssh
csrc/adam/compat.h
csrc/adam/cpu_adam.cpp
csrc/adam/custom_cuda_kernel.cu
csrc/adam/fused_adam_frontend.cpp
csrc/adam/multi_tensor_adam.cu
csrc/adam/multi_tensor_apply.cuh
csrc/aio/common/deepspeed_aio_common.cpp
csrc/aio/common/deepspeed_aio_common.h
csrc/aio/common/deepspeed_aio_types.cpp
csrc/aio/common/deepspeed_aio_types.h
csrc/aio/common/deepspeed_aio_utils.cpp
csrc/aio/common/deepspeed_aio_utils.h
csrc/aio/py_lib/deepspeed_aio_thread.cpp
csrc/aio/py_lib/deepspeed_aio_thread.h
csrc/aio/py_lib/deepspeed_py_aio.cpp
csrc/aio/py_lib/deepspeed_py_aio.h
csrc/aio/py_lib/deepspeed_py_aio_handle.cpp
csrc/aio/py_lib/deepspeed_py_aio_handle.h
csrc/aio/py_lib/deepspeed_py_copy.cpp
csrc/aio/py_lib/deepspeed_py_copy.h
csrc/aio/py_lib/py_ds_aio.cpp
csrc/includes/StopWatch.h
csrc/includes/Timer.h
csrc/includes/context.h
csrc/includes/cpu_adam.h
csrc/includes/cublas_wrappers.h
csrc/includes/custom_cuda_layers.h
csrc/includes/dropout.h
csrc/includes/ds_transformer_cuda.h
csrc/includes/feed_forward.h
csrc/includes/gelu.h
csrc/includes/gemm_test.h
csrc/includes/general_kernels.h
csrc/includes/normalize_layer.h
csrc/includes/quantizer.h
csrc/includes/softmax.h
csrc/includes/strided_batch_gemm.h
csrc/includes/type_shim.h
csrc/lamb/fused_lamb_cuda.cpp
csrc/lamb/fused_lamb_cuda_kernel.cu
csrc/quantization/pt_binding.cpp
csrc/quantization/quantizer.cu
csrc/sparse_attention/utils.cpp
csrc/transformer/cublas_wrappers.cu
csrc/transformer/dropout_kernels.cu
csrc/transformer/ds_transformer_cuda.cpp
csrc/transformer/gelu_kernels.cu
csrc/transformer/general_kernels.cu
csrc/transformer/normalize_kernels.cu
csrc/transformer/softmax_kernels.cu
csrc/transformer/transform_kernels.cu
csrc/transformer/inference/csrc/dequantize.cu
csrc/transformer/inference/csrc/gelu.cu
csrc/transformer/inference/csrc/normalize.cu
csrc/transformer/inference/csrc/pt_binding.cpp
csrc/transformer/inference/csrc/softmax.cu
csrc/transformer/inference/includes/context.h
csrc/transformer/inference/includes/cublas_wrappers.h
csrc/transformer/inference/includes/custom_cuda_layers.h
csrc/utils/flatten_unflatten.cpp
deepspeed/__init__.py
deepspeed/constants.py
deepspeed/env_report.py
deepspeed/git_version_info.py
deepspeed/git_version_info_installed.py
deepspeed.egg-info/PKG-INFO
deepspeed.egg-info/SOURCES.txt
deepspeed.egg-info/dependency_links.txt
deepspeed.egg-info/entry_points.txt
deepspeed.egg-info/requires.txt
deepspeed.egg-info/top_level.txt
deepspeed/elasticity/__init__.py
deepspeed/elasticity/config.py
deepspeed/elasticity/constants.py
deepspeed/elasticity/elasticity.py
deepspeed/inference/__init__.py
deepspeed/inference/engine.py
deepspeed/launcher/__init__.py
deepspeed/launcher/constants.py
deepspeed/launcher/launch.py
deepspeed/launcher/multinode_runner.py
deepspeed/launcher/runner.py
deepspeed/module_inject/__init__.py
deepspeed/module_inject/inject.py
deepspeed/module_inject/module_quantize.py
deepspeed/module_inject/replace_module.py
deepspeed/module_inject/replace_policy.py
deepspeed/ops/__init__.py
deepspeed/ops/adam/__init__.py
deepspeed/ops/adam/cpu_adam.py
deepspeed/ops/adam/fused_adam.py
deepspeed/ops/adam/multi_tensor_apply.py
deepspeed/ops/aio/__init__.py
deepspeed/ops/csrc/adam/compat.h
deepspeed/ops/csrc/adam/cpu_adam.cpp
deepspeed/ops/csrc/adam/custom_cuda_kernel.cu
deepspeed/ops/csrc/adam/fused_adam_frontend.cpp
deepspeed/ops/csrc/adam/multi_tensor_adam.cu
deepspeed/ops/csrc/adam/multi_tensor_apply.cuh
deepspeed/ops/csrc/aio/common/deepspeed_aio_common.cpp
deepspeed/ops/csrc/aio/common/deepspeed_aio_common.h
deepspeed/ops/csrc/aio/common/deepspeed_aio_types.cpp
deepspeed/ops/csrc/aio/common/deepspeed_aio_types.h
deepspeed/ops/csrc/aio/common/deepspeed_aio_utils.cpp
deepspeed/ops/csrc/aio/common/deepspeed_aio_utils.h
deepspeed/ops/csrc/aio/py_lib/deepspeed_aio_thread.cpp
deepspeed/ops/csrc/aio/py_lib/deepspeed_aio_thread.h
deepspeed/ops/csrc/aio/py_lib/deepspeed_py_aio.cpp
deepspeed/ops/csrc/aio/py_lib/deepspeed_py_aio.h
deepspeed/ops/csrc/aio/py_lib/deepspeed_py_aio_handle.cpp
deepspeed/ops/csrc/aio/py_lib/deepspeed_py_aio_handle.h
deepspeed/ops/csrc/aio/py_lib/deepspeed_py_copy.cpp
deepspeed/ops/csrc/aio/py_lib/deepspeed_py_copy.h
deepspeed/ops/csrc/aio/py_lib/py_ds_aio.cpp
deepspeed/ops/csrc/includes/StopWatch.h
deepspeed/ops/csrc/includes/Timer.h
deepspeed/ops/csrc/includes/context.h
deepspeed/ops/csrc/includes/cpu_adam.h
deepspeed/ops/csrc/includes/cublas_wrappers.h
deepspeed/ops/csrc/includes/custom_cuda_layers.h
deepspeed/ops/csrc/includes/dropout.h
deepspeed/ops/csrc/includes/ds_transformer_cuda.h
deepspeed/ops/csrc/includes/feed_forward.h
deepspeed/ops/csrc/includes/gelu.h
deepspeed/ops/csrc/includes/gemm_test.h
deepspeed/ops/csrc/includes/general_kernels.h
deepspeed/ops/csrc/includes/normalize_layer.h
deepspeed/ops/csrc/includes/quantizer.h
deepspeed/ops/csrc/includes/softmax.h
deepspeed/ops/csrc/includes/strided_batch_gemm.h
deepspeed/ops/csrc/includes/type_shim.h
deepspeed/ops/csrc/lamb/fused_lamb_cuda.cpp
deepspeed/ops/csrc/lamb/fused_lamb_cuda_kernel.cu
deepspeed/ops/csrc/quantization/pt_binding.cpp
deepspeed/ops/csrc/quantization/quantizer.cu
deepspeed/ops/csrc/sparse_attention/utils.cpp
deepspeed/ops/csrc/transformer/cublas_wrappers.cu
deepspeed/ops/csrc/transformer/dropout_kernels.cu
deepspeed/ops/csrc/transformer/ds_transformer_cuda.cpp
deepspeed/ops/csrc/transformer/gelu_kernels.cu
deepspeed/ops/csrc/transformer/general_kernels.cu
deepspeed/ops/csrc/transformer/normalize_kernels.cu
deepspeed/ops/csrc/transformer/softmax_kernels.cu
deepspeed/ops/csrc/transformer/transform_kernels.cu
deepspeed/ops/csrc/transformer/inference/csrc/dequantize.cu
deepspeed/ops/csrc/transformer/inference/csrc/gelu.cu
deepspeed/ops/csrc/transformer/inference/csrc/normalize.cu
deepspeed/ops/csrc/transformer/inference/csrc/pt_binding.cpp
deepspeed/ops/csrc/transformer/inference/csrc/softmax.cu
deepspeed/ops/csrc/transformer/inference/includes/context.h
deepspeed/ops/csrc/transformer/inference/includes/cublas_wrappers.h
deepspeed/ops/csrc/transformer/inference/includes/custom_cuda_layers.h
deepspeed/ops/csrc/utils/flatten_unflatten.cpp
deepspeed/ops/lamb/__init__.py
deepspeed/ops/lamb/fused_lamb.py
deepspeed/ops/op_builder/__init__.py
deepspeed/ops/op_builder/async_io.py
deepspeed/ops/op_builder/builder.py
deepspeed/ops/op_builder/cpu_adam.py
deepspeed/ops/op_builder/fused_adam.py
deepspeed/ops/op_builder/fused_lamb.py
deepspeed/ops/op_builder/quantizer.py
deepspeed/ops/op_builder/sparse_attn.py
deepspeed/ops/op_builder/stochastic_transformer.py
deepspeed/ops/op_builder/transformer.py
deepspeed/ops/op_builder/transformer_inference.py
deepspeed/ops/op_builder/utils.py
deepspeed/ops/quantizer/__init__.py
deepspeed/ops/quantizer/quantizer.py
deepspeed/ops/sparse_attention/__init__.py
deepspeed/ops/sparse_attention/bert_sparse_self_attention.py
deepspeed/ops/sparse_attention/matmul.py
deepspeed/ops/sparse_attention/softmax.py
deepspeed/ops/sparse_attention/sparse_attention_utils.py
deepspeed/ops/sparse_attention/sparse_self_attention.py
deepspeed/ops/sparse_attention/sparsity_config.py
deepspeed/ops/sparse_attention/trsrc/__init__.py
deepspeed/ops/sparse_attention/trsrc/matmul.tr
deepspeed/ops/sparse_attention/trsrc/softmax_bwd.tr
deepspeed/ops/sparse_attention/trsrc/softmax_fwd.tr
deepspeed/ops/transformer/__init__.py
deepspeed/ops/transformer/transformer.py
deepspeed/ops/transformer/inference/__init__.py
deepspeed/ops/transformer/inference/transformer_inference.py
deepspeed/pipe/__init__.py
deepspeed/profiling/__init__.py
deepspeed/profiling/config.py
deepspeed/profiling/constants.py
deepspeed/profiling/flops_profiler/__init__.py
deepspeed/profiling/flops_profiler/profiler.py
deepspeed/runtime/__init__.py
deepspeed/runtime/config.py
deepspeed/runtime/config_utils.py
deepspeed/runtime/constants.py
deepspeed/runtime/csr_tensor.py
deepspeed/runtime/dataloader.py
deepspeed/runtime/eigenvalue.py
deepspeed/runtime/engine.py
deepspeed/runtime/lr_schedules.py
deepspeed/runtime/progressive_layer_drop.py
deepspeed/runtime/quantize.py
deepspeed/runtime/state_dict_factory.py
deepspeed/runtime/utils.py
deepspeed/runtime/weight_quantizer.py
deepspeed/runtime/activation_checkpointing/__init__.py
deepspeed/runtime/activation_checkpointing/checkpointing.py
deepspeed/runtime/activation_checkpointing/config.py
deepspeed/runtime/comm/__init__.py
deepspeed/runtime/comm/mpi.py
deepspeed/runtime/comm/nccl.py
deepspeed/runtime/compression/__init__.py
deepspeed/runtime/compression/cupy.py
deepspeed/runtime/fp16/__init__.py
deepspeed/runtime/fp16/fused_optimizer.py
deepspeed/runtime/fp16/loss_scaler.py
deepspeed/runtime/fp16/unfused_optimizer.py
deepspeed/runtime/fp16/onebit/__init__.py
deepspeed/runtime/fp16/onebit/adam.py
deepspeed/runtime/fp16/onebit/lamb.py
deepspeed/runtime/pipe/__init__.py
deepspeed/runtime/pipe/engine.py
deepspeed/runtime/pipe/module.py
deepspeed/runtime/pipe/p2p.py
deepspeed/runtime/pipe/schedule.py
deepspeed/runtime/pipe/topology.py
deepspeed/runtime/swap_tensor/__init__.py
deepspeed/runtime/swap_tensor/aio_config.py
deepspeed/runtime/swap_tensor/async_swapper.py
deepspeed/runtime/swap_tensor/constants.py
deepspeed/runtime/swap_tensor/optimizer_utils.py
deepspeed/runtime/swap_tensor/partitioned_optimizer_swapper.py
deepspeed/runtime/swap_tensor/partitioned_param_swapper.py
deepspeed/runtime/swap_tensor/pipelined_optimizer_swapper.py
deepspeed/runtime/swap_tensor/utils.py
deepspeed/runtime/zero/__init__.py
deepspeed/runtime/zero/config.py
deepspeed/runtime/zero/constants.py
deepspeed/runtime/zero/contiguous_memory_allocator.py
deepspeed/runtime/zero/linear.py
deepspeed/runtime/zero/offload_config.py
deepspeed/runtime/zero/offload_constants.py
deepspeed/runtime/zero/partition_parameters.py
deepspeed/runtime/zero/stage1.py
deepspeed/runtime/zero/stage2.py
deepspeed/runtime/zero/stage3.py
deepspeed/runtime/zero/test.py
deepspeed/runtime/zero/tiling.py
deepspeed/runtime/zero/utils.py
deepspeed/utils/__init__.py
deepspeed/utils/debug.py
deepspeed/utils/distributed.py
deepspeed/utils/logging.py
deepspeed/utils/timer.py
deepspeed/utils/zero_to_fp32.py
op_builder/__init__.py
op_builder/async_io.py
op_builder/builder.py
op_builder/cpu_adam.py
op_builder/fused_adam.py
op_builder/fused_lamb.py
op_builder/quantizer.py
op_builder/sparse_attn.py
op_builder/stochastic_transformer.py
op_builder/transformer.py
op_builder/transformer_inference.py
op_builder/utils.py
requirements/requirements-1bit-adam.txt
requirements/requirements-dev.txt
requirements/requirements-readthedocs.txt
requirements/requirements-sparse_attn.txt
requirements/requirements.txt