MCPcopy Create free account

hub / github.com/bytedance/ABQ-LLM / functions

Functions14,558 in github.com/bytedance/ABQ-LLM

⨍ Functions 14,558◇ Types & classes 4,979
↓ 1,736 callersMethodappend
Appends an operation and takes ownership
fastertransformer/3rdparty/cutlass/tools/library/src/manifest.cpp:81
↓ 1,684 callersMethodat
prevent converting int or size_t to string automatically
fastertransformer/src/fastertransformer/utils/Tensor.h:358
↓ 969 callersMethodsize
fastertransformer/src/fastertransformer/utils/Tensor.cc:212
↓ 890 callersMethodfind
Returns the cached result
fastertransformer/3rdparty/cutlass/test/unit/conv/device/cache_testbed_output.h:178
↓ 811 callersMethodhost_view
Accesses the tensor reference pointing to data
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:288
↓ 776 callersMethodat
Accesses an element by coordinate
fastertransformer/3rdparty/cutlass/include/cutlass/matrix.h:148
↓ 762 callersMethodreMalloc
fastertransformer/src/fastertransformer/utils/allocator.h:75
↓ 696 callersMethodfree
fastertransformer/src/fastertransformer/utils/allocator.h:348
↓ 548 callersMethodinsert
fastertransformer/src/fastertransformer/utils/Tensor.h:338
↓ 515 callersMethodcolumn
fastertransformer/3rdparty/cutlass/include/cutlass/matrix.h:2094
↓ 493 callersMethodrow
fastertransformer/3rdparty/cutlass/include/cutlass/matrix.h:229
↓ 465 callersMethodGetInteger
fastertransformer/3rdparty/INIReader.h:411
↓ 409 callersMethodinfo
(self, s)
fastertransformer/examples/pytorch/vit/ViT-quantization/quant_utils.py:30
↓ 392 callersMethodcuda
(self)
fastertransformer/examples/pytorch/gpt/utils/gpt.py:550
↓ 377 callersMethoddata
Returns the pointer to referenced data
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_ref.h:268
↓ 342 callersMethodview
Returns a TensorRef pointing to the first element of the tensor.
fastertransformer/3rdparty/cutlass/include/cutlass/thread/matrix.h:153
↓ 309 callersMethodto
(self, device=None)
fastertransformer/examples/pytorch/gpt/utils/gpt_decoder.py:954
↓ 291 callersMethodget
fastertransformer/3rdparty/cutlass/include/cutlass/array_subbyte.h:125
↓ 273 callersMethodsync_device
Copies data from host to device
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:377
↓ 266 callersMethoddata
fastertransformer/examples/cpp/xlnet/cnpy.h:36
↓ 252 callersFunctionto_string
fastertransformer/3rdparty/cutlass/test/unit/gemm/device/testbed_utils.h:39
↓ 251 callersMethodtranspose
fastertransformer/3rdparty/cutlass/include/cutlass/matrix.h:138
↓ 247 callersMethoddevice_ref
Accesses the tensor reference pointing to data
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:278
↓ 237 callersMethoddata
Returns a pointer to the shared memory buffer
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/epilogue_base.h:179
↓ 236 callersMethodstride
Returns the stride of the layout
fastertransformer/3rdparty/cutlass/include/cutlass/layout/matrix.h:118
↓ 229 callersFunctionat
()
fastertransformer/3rdparty/cutlass/docs/jquery.js:23
↓ 222 callersMethodend
fastertransformer/src/fastertransformer/utils/Tensor.h:501
↓ 206 callersMethodat
Returns a reference to the element at the logical Coord in host memory
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:349
↓ 205 callersFunctionprint_rank_0
If distributed is initialized, print only on rank 0.
fastertransformer/3rdparty/Megatron-LM/megatron/__init__.py:28
↓ 198 callersMethodones
Constructs a matrix from a uniform element 1
fastertransformer/3rdparty/cutlass/include/cutlass/matrix.h:125
↓ 197 callersMethodnumpy
(x: np.ndarray)
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:322
↓ 196 callersMethodat
Gets the value pointed to
fastertransformer/3rdparty/cutlass/tools/profiler/src/problem_space.cpp:129
↓ 193 callersMethodget_tensor
fastertransformer/src/fastertransformer/tf_op/BaseOp.h:68
↓ 192 callersFunctioninitialize
fastertransformer/src/fastertransformer/utils/mpi_utils.cc:37
↓ 191 callersFunctionElement
Unpacks an element from memory
fastertransformer/3rdparty/cutlass/include/cutlass/subbyte_reference.h:147
↓ 191 callersMethodbegin
fastertransformer/src/fastertransformer/utils/Tensor.h:496
↓ 179 callersMethodempty
fastertransformer/3rdparty/cutlass/include/cutlass/array.h:427
↓ 175 callersMethodmn
Obtains a Coord<2> from GemmCoord
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/gemm.h:175
↓ 166 callersMethodwrite
Write timers to a tensorboard writer
fastertransformer/3rdparty/Megatron-LM/megatron/global_vars.py:252
↓ 163 callersMethodload
Loads a fragment from memory
fastertransformer/3rdparty/cutlass/examples/42_fused_multi_head_attention/iterators/predicated_tile_iterator_residual_last.h:887
↓ 161 callersMethodstride
Returns the layout object's stride vector
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:324
↓ 150 callersFunctionget_args
Return arguments.
fastertransformer/3rdparty/Megatron-LM/megatron/global_vars.py:38
↓ 148 callersMethodadd_tile_offset
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/shared_load_iterator_mixed.h:184
↓ 142 callersMethodreset
Clears the HostTensor allocation to size/capacity = 0
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:157
↓ 139 callersMethodadd_tile_offset
Advances an iterator along logical dimensions of matrix in units of whole tiles
fastertransformer/3rdparty/cutlass/examples/42_fused_multi_head_attention/iterators/predicated_tile_access_iterator_residual_last.h:816
↓ 139 callersMethodhost_ref
Accesses the tensor reference pointing to data
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:272
↓ 137 callersFunctionadd_tile_offset
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/shared_load_iterator.h:155
↓ 129 callersMethodclear
Efficient clear method
fastertransformer/3rdparty/cutlass/include/cutlass/array.h:361
↓ 127 callersMethodstr
fastertransformer/3rdparty/cutlass/test/unit/conv/device/cache_testbed_output.h:217
↓ 125 callersMethodw
Accesses the real part of the quaternion
fastertransformer/3rdparty/cutlass/include/cutlass/quaternion.h:187
↓ 123 callersMethoddevice
(self)
fastertransformer/examples/pytorch/gpt/utils/gpt_decoder.py:944
↓ 123 callersMethoddevice_data
Gets pointer to device data
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:263
↓ 121 callersMethodGemm
fastertransformer/src/fastertransformer/utils/gemm.cc:23
↓ 120 callersMethodcount
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/predicated_tile_iterator_params.h:81
↓ 119 callersMethodcapacity
Returns the logical capacity based on extent and layout. May differ from size().
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:239
↓ 117 callersMethodset_kgroup_index
Notify the iterator which k-group it is currently pointing to. This does not advance the iterator. Rather, it overrides its internal tracking with co
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/warp/tile_iterator_planar_complex.h:238
↓ 115 callersMethodappend
Inserts the operation. operation_kind -> configuration_name -> []
fastertransformer/3rdparty/cutlass/tools/library/scripts/manifest.py:322
↓ 115 callersMethodinfo
(self, s)
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/quant_utils.py:31
↓ 113 callersMethodadd_pointer_offset
Adds a pointer offset in units of Element
fastertransformer/3rdparty/cutlass/include/cutlass/conv/threadblock/conv2d_tile_iterator.h:114
↓ 113 callersMethodisExist
fastertransformer/src/fastertransformer/utils/cublasAlgoMap.cc:133
↓ 113 callersMethodload
(self, ckpt_path)
fastertransformer/examples/pytorch/gpt/utils/gpt.py:539
↓ 112 callersMethodcast
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/reference/host/gemm.h:53
↓ 112 callersFunctionenum_auto
()
fastertransformer/3rdparty/cutlass/tools/library/scripts/library.py:22
↓ 112 callersMethodkeys
(self)
fastertransformer/examples/pytorch/gptj/utils/reference_gptj.py:36
↓ 110 callersMethodget
(self)
fastertransformer/3rdparty/cutlass/tools/library/scripts/rt.py:610
↓ 110 callersMethodresize
Changes the size of the logical tensor. Only allocates memory if new capacity exceeds reserved capacity. To force allocation, call reset().
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:209
↓ 109 callersMethodValue
fastertransformer/3rdparty/cutlass/tools/profiler/src/problem_space.h:125
↓ 109 callersMethodget
fastertransformer/src/fastertransformer/tensorrt_plugin/bert_fp8/bertFp8Plugin.h:78
↓ 108 callersMethodextent
Returns the extent of the tensor
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:359
↓ 107 callersMethodexists
(path)
fastertransformer/3rdparty/Megatron-LM/megatron/data/indexed_dataset.py:201
↓ 106 callersMethodTensor4DCoord
Default ctor
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_coord.h:72
↓ 105 callersFunctionTensorNorm
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/reference/host/tensor_reduce.h:154
↓ 104 callersMethodinfo
(self, s)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/quant_utils.py:30
↓ 103 callersMethodset_iteration_index
Overrides the internal iteration index
fastertransformer/3rdparty/cutlass/include/cutlass/conv/threadblock/conv2d_tile_iterator.h:108
↓ 102 callersMethodcuda
(self)
fastertransformer/examples/pytorch/baichuan2/baichuanv2.py:262
↓ 99 callersMethodGetFloat
fastertransformer/3rdparty/INIReader.h:445
↓ 97 callersMethodclear
< Efficiently disables all accesses guarded by mask
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/predicated_tile_iterator.h:153
↓ 96 callersMethodGet
fastertransformer/3rdparty/INIReader.h:394
↓ 94 callersMethodapply
Compute Softmax
fastertransformer/3rdparty/cutlass/examples/35_gemm_softmax/gemm_with_softmax.h:199
↓ 93 callersFunctionTensorFill
< Layout function
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/reference/host/tensor_fill.h:122
↓ 91 callersMethodfloat
(self)
fastertransformer/examples/pytorch/gpt/utils/gpt_decoder.py:959
↓ 90 callersMethodget
Returns a pointer
fastertransformer/3rdparty/cutlass/examples/42_fused_multi_head_attention/iterators/predicated_tile_access_iterator_residual_last.h:822
↓ 90 callersMethodset_iteration_index
Overrides the internal iteration index
fastertransformer/3rdparty/cutlass/examples/42_fused_multi_head_attention/iterators/predicated_tile_access_iterator_residual_last.h:798
↓ 89 callersMethodclear_mask
< Efficiently disables all accesses guarded by mask
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/predicated_tile_iterator_v2.h:434
↓ 89 callersMethodresize
Changes the size of the view without affecting pointer or layout
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_view.h:170
↓ 89 callersMethodsum
Returns the sum of all elements
fastertransformer/3rdparty/cutlass/include/cutlass/coord.h:301
↓ 88 callersMethodclose
(self)
fastertransformer/examples/tensorflow/bert/tensorflow_bert/bert/run_squad.py:1093
↓ 85 callersMethodoffset
Computes the offset of an index from the origin of the tensor
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_ref.h:315
↓ 85 callersMethodsync_host
Copies data from device to host
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:369
↓ 84 callersMethodsize
(self, index)
fastertransformer/3rdparty/Megatron-LM/megatron/data/indexed_dataset.py:197
↓ 80 callersMethodhalf
(self)
fastertransformer/examples/pytorch/llama/llamav1.py:250
↓ 79 callersFunctionfmtstr
fastertransformer/src/fastertransformer/utils/string_utils.h:27
↓ 77 callersFunctionproduct
(X, identity = 1)
fastertransformer/3rdparty/cutlass/tools/library/scripts/generator.py:39
↓ 77 callersFunctionsum
fastertransformer/src/fastertransformer/kernels/decoder_masked_multihead_attention_utils.h:1188
↓ 76 callersFunctionb
(e)
fastertransformer/3rdparty/cutlass/docs/jquery.js:32
↓ 75 callersMethodlayout
Returns the layout object
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/host_tensor.h:319
↓ 75 callersFunctionrun
()
fastertransformer/examples/pytorch/bert/bert_example.py:287
↓ 74 callersMethodMatrixCoord
Default ctor
fastertransformer/3rdparty/cutlass/include/cutlass/matrix_coord.h:73
↓ 74 callersMethodload
Loads a fragment from memory
fastertransformer/3rdparty/cutlass/include/cutlass/conv/threadblock/conv2d_tile_iterator.h:166
↓ 74 callersMethodread
(path)
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/SwinTransformer/data/zipreader.py:86
next →1–100 of 14,558, ranked by callers