MCPcopy Create free account

hub / github.com/bytedance/ABQ-LLM / types & classes

Types & classes4,979 in github.com/bytedance/ABQ-LLM

⨍ Functions 14,558◇ Types & classes 4,979
↓ 441 callersClassTileDescription
fastertransformer/3rdparty/cutlass/tools/library/scripts/library.py:667
↓ 240 callersClassMatrixCoord
MatrixCoord wraps Coord<2, int> to provide a helper for accessing named dimensions. Classes expecting a coordinate in the rank=2 index space of a matr
fastertransformer/3rdparty/cutlass/include/cutlass/matrix_coord.h:45
↓ 222 callersClassTensorDescription
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/library.py:687
↓ 126 callersClassPitchLinearCoord
Coordinate in pitch-linear space
fastertransformer/3rdparty/cutlass/include/cutlass/pitch_linear_coord.h:57
↓ 106 callersClassMatrix
fastertransformer/3rdparty/cutlass/include/cutlass/matrix.h:54
↓ 84 callersClassMathInstruction
fastertransformer/3rdparty/cutlass/tools/library/scripts/library.py:657
↓ 84 callersClassMathInstruction
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/library.py:652
↓ 74 callersClassTileDescription
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/library.py:664
↓ 69 callersClassLinearCombination
Apply a linear combination operator to an array of elements D = alpha * accumulator + beta * source :param element_output: data type use
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:85
↓ 69 callersClasshalf_t
IEEE half-precision floating-point type
fastertransformer/3rdparty/cutlass/include/cutlass/half.h:183
↓ 68 callersClassConv2dProblemSize
fastertransformer/3rdparty/cutlass/include/cutlass/conv/conv2d_problem_size.h:71
↓ 51 callersClassTensor
fastertransformer/src/fastertransformer/utils/Tensor.h:112
↓ 48 callersClasscublasMMWrapper
fastertransformer/src/fastertransformer/utils/cublasMMWrapper.h:30
↓ 47 callersClassTensorDescription
fastertransformer/3rdparty/cutlass/tools/library/scripts/library.py:681
↓ 46 callersClassAverageMeter
Computes and stores the average and current value
fastertransformer/examples/pytorch/vit/ViT-quantization/main.py:46
↓ 46 callersClassIndex
fastertransformer/3rdparty/Megatron-LM/megatron/data/indexed_dataset.py:336
↓ 45 callersClassConv2dOperation
CUTLASS Conv2d operation description. :param conv_kind: convolution operator :type conv_kind: :class:`cutlass.conv.Operator` :param
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/conv2d_operation.py:357
↓ 38 callersClassbfloat16_t
Floating-point type with 8 bits of exponent and 7 bits of mantissa.
fastertransformer/3rdparty/cutlass/include/cutlass/bfloat16.h:52
↓ 37 callersClasstfloat32_t
Tensor Float 32 data type
fastertransformer/3rdparty/cutlass/include/cutlass/tfloat32.h:50
↓ 35 callersClassINIReader
Read an INI file into easy-to-access name/value pairs. (Note that I've gone for simplicity here rather than speed, but it should be pretty decent.)
fastertransformer/3rdparty/INIReader.h:306
↓ 30 callersClassPitchLinearCoord
fastertransformer/3rdparty/cutlass/tools/library/scripts/rt.py:87
↓ 26 callersClassGemmOperationUniversal
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/gemm_operation.py:936
↓ 25 callersClassPitchLinear
Mapping function for pitch-linear memory
fastertransformer/3rdparty/cutlass/include/cutlass/layout/pitch_linear.h:50
↓ 25 callersClassSubPolicy
Definition of a SubPolicy. A SubPolicy consists of two augmentation operations, each of those parametrized as operation, probability, magnitu
fastertransformer/3rdparty/Megatron-LM/megatron/data/autoaugment.py:113
↓ 24 callersClassTensorMap
fastertransformer/src/fastertransformer/utils/Tensor.h:309
↓ 24 callersClassTensorView
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_view.h:62
↓ 23 callersClassGemmCoord
GemmCoord is a structure derived from Coord<3> that specifies a location within the coordinate space of a GEMM problem.
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/gemm.h:99
↓ 21 callersClassElement
Structure representing an element in source memory
fastertransformer/3rdparty/cutlass/examples/03_visualize_layout/visualize_layout.h:130
↓ 21 callersEnumPositionEmbeddingType
fastertransformer/src/fastertransformer/kernels/gen_relative_pos_bias.h:27
↓ 18 callersClassConv3dProblemSize
fastertransformer/3rdparty/cutlass/include/cutlass/conv/conv3d_problem_size.h:59
↓ 18 callersClassQuantLinear
Quantized Module that can perform quantized convolution or normal convolution. To activate quantization, please use set_quant_state function.
algorithm/quantize/int_linear.py:25
↓ 17 callersClassGemmCoord
fastertransformer/3rdparty/cutlass/tools/library/scripts/rt.py:93
↓ 17 callersClassKeyHandler
fastertransformer/examples/pytorch/gptneox/utils/eleutherai_gpt_neox_convert.py:86
↓ 16 callersClassParams
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/kernel/gemm.h:72
↓ 16 callersClassTensorDescription
Structure describing the properties of a tensor
fastertransformer/3rdparty/cutlass/tools/library/include/cutlass/library/library.h:423
↓ 15 callersClassParams
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/reference/device/tensor_fill.h:412
↓ 14 callersClassMask
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/predicated_tile_iterator.h:137
↓ 13 callersClassFakeQuantizer
Fake Tensor quantizer module This module quantize a tensor and wraps variable. It also can collect relevant variables for calibration. Args:
fastertransformer/examples/tensorflow/bert/bert-quantization/ft-tensorflow-quantization/ft_tensorflow_quantization/python/layers/tensor_quantizer.py:126
↓ 13 callersClassGemmNotSupportedException
fastertransformer/src/fastertransformer/utils/gemm.h:571
↓ 13 callersClassParams
fastertransformer/3rdparty/cutlass/include/cutlass/conv/threadblock/conv3d_wgrad_activation_tile_access_iterator_analytic.h:97
↓ 12 callersClassLayerNorm
Apply a linear combination operator to an array of elements D = alpha * accumulator + beta * source :param element_output: data type use
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:453
↓ 12 callersClassQuantDescriptor
Supportive descriptor of quantization. Describe how a tensor should be quantized. Args: collection_name_prefix: A string. Determine which colle
fastertransformer/examples/tensorflow/bert/bert-quantization/ft-tensorflow-quantization/ft_tensorflow_quantization/python/layers/tensor_quantizer.py:28
↓ 12 callersClassTensor4DCoord
Defines a canonical 4D coordinate used by tensor operations.
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_coord.h:44
↓ 12 callersClassrelu
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:305
↓ 11 callersClassTensorRef
fastertransformer/3rdparty/cutlass/tools/library/scripts/rt.py:126
↓ 11 callersClassidentity
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:300
↓ 10 callersClassLinearActivation
r"""Fused Linear and Activation Module.
fastertransformer/examples/pytorch/swin/Swin-Transformer-Quantization/models/linear_activation.py:91
↓ 10 callersClassTensor5DCoord
Defines a canonical 5D coordinate used by tensor operations.
fastertransformer/3rdparty/cutlass/include/cutlass/tensor_coord.h:180
↓ 9 callersClassParams
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/kernel/rank_2k_grouped.h:286
↓ 9 callersClassVisionTransformerINT8
fastertransformer/examples/pytorch/vit/ViT-quantization/vit_int8.py:399
↓ 8 callersClassConv2dOperation
fastertransformer/3rdparty/cutlass/tools/library/scripts/conv2d_operation.py:17
↓ 8 callersClassEpilogueTileDesc
fastertransformer/3rdparty/cutlass/tools/library/scripts/rt.py:206
↓ 8 callersClassFastDivmod
Object to encapsulate the fast division+modulus operation. This object precomputes two values used to accelerate the computation and is best used whe
fastertransformer/3rdparty/cutlass/include/cutlass/fast_math.h:277
↓ 8 callersClassLinearActivation
r"""Fused Linear and activation Module.
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/modeling.py:150
↓ 8 callersClassPredicatedTileIterator
fastertransformer/3rdparty/cutlass/include/cutlass/transform/threadblock/predicated_tile_iterator.h:142
↓ 8 callersClassPredicatedTileIteratorResidualLast
fastertransformer/3rdparty/cutlass/examples/42_fused_multi_head_attention/iterators/predicated_tile_iterator_residual_last.h:151
↓ 8 callersClassTranslationResult
fastertransformer/examples/tensorflow/t5/translate_example.py:67
↓ 8 callersClassTranslationResult
fastertransformer/examples/pytorch/bart/translate_example.py:48
↓ 8 callersClassTranslationResult
fastertransformer/examples/pytorch/t5/translate_example.py:81
↓ 8 callersClassTranslationResult
fastertransformer/examples/pytorch/t5/perf_benchmark.py:46
↓ 8 callersClassUnpackedNemoCheckpointDir
fastertransformer/examples/pytorch/nemo.py:76
↓ 7 callersClassBertModel
BERT model ("Bidirectional Embedding Representations from a Transformer"). Params: config: a BertConfig class instance with the configura
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/modeling.py:855
↓ 7 callersClassTensor
fastertransformer/src/fastertransformer/triton_backend/transformer_triton_backend.hpp:101
↓ 7 callersClassTranslationResult
fastertransformer/examples/tensorflow/decoding/translate_example.py:266
↓ 6 callersClassAbqLayerNorm
algorithm/quantize/abq_norm.py:25
↓ 6 callersClassBatchedGemmCoord
BatchedGemmCoord is a structure derived from Coord<4> that specifies a location within the coordinate space of a batched GEMM problem.
fastertransformer/3rdparty/cutlass/include/cutlass/gemm/gemm.h:265
↓ 6 callersClassBlendableDataset
fastertransformer/3rdparty/Megatron-LM/megatron/data/blendable_dataset.py:27
↓ 6 callersClassGemmOperationGrouped
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/gemm_operation.py:946
↓ 6 callersClassLinearActivation
r"""Fused Linear and Activation Module.
fastertransformer/examples/pytorch/vit/ViT-quantization/vit_int8.py:80
↓ 6 callersClassParams
fastertransformer/3rdparty/cutlass/tools/util/include/cutlass/util/reference/device/tensor_relu.h:69
↓ 6 callersClassPreActBottleneck
Pre-activation (v2) bottleneck block.
fastertransformer/examples/pytorch/vit/ViT-quantization/ViT-pytorch/models/modeling_resnet.py:55
↓ 6 callersClassPredicatedTileAccessIterator
fastertransformer/3rdparty/cutlass/include/cutlass/transform/threadblock/predicated_tile_access_iterator.h:315
↓ 6 callersClassPredicatedTileAccessIteratorResidualLast
fastertransformer/3rdparty/cutlass/examples/42_fused_multi_head_attention/iterators/predicated_tile_access_iterator_residual_last.h:79
↓ 6 callersClasstanh
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:365
↓ 6 callersClassuint128_t
! Unsigned 128b integer type
fastertransformer/3rdparty/cutlass/include/cutlass/uint128.h:71
↓ 5 callersClassAverageMeter
Computes and stores the average and current value
fastertransformer/examples/pytorch/vit/ViT-quantization/eval_engine.py:48
↓ 5 callersClassConv3dOperation
fastertransformer/3rdparty/cutlass/tools/library/scripts/conv3d_operation.py:17
↓ 5 callersClassDecodingArgumentNew
fastertransformer/examples/tensorflow/decoder/utils/common.py:214
↓ 5 callersClassDecodingWeights
fastertransformer/examples/pytorch/decoding/utils/decoding.py:49
↓ 5 callersClassEthicsConfig
BuilderConfig for Hendrycks ETHICS.
algorithm/lm_eval/datasets/hendrycks_ethics/hendrycks_ethics.py:50
↓ 5 callersClassFTT5
fastertransformer/examples/pytorch/t5/utils/ft_decoding.py:488
↓ 5 callersClassFTT5Decoding
fastertransformer/examples/pytorch/t5/utils/ft_decoding.py:426
↓ 5 callersClassFTT5DecodingWeight
fastertransformer/examples/pytorch/t5/utils/ft_decoding.py:22
↓ 5 callersClassFTT5Encoder
fastertransformer/examples/pytorch/t5/utils/ft_encoder.py:360
↓ 5 callersClassFTT5EncoderWeight
fastertransformer/examples/pytorch/t5/utils/ft_encoder.py:23
↓ 5 callersClassFusedScaleMaskSoftmax
fused operation: scaling + mask + softmax Arguments: input_in_fp16: flag to indicate if input in fp16 data format. input_in_
fastertransformer/3rdparty/Megatron-LM/megatron/model/fused_softmax.py:115
↓ 5 callersClassGPTModel
GPT-2 Language model.
fastertransformer/3rdparty/Megatron-LM/megatron/model/gpt_model.py:52
↓ 5 callersClassInputExample
A single training/test example for simple sequence classification.
fastertransformer/examples/tensorflow/bert/tensorflow_bert/bert/run_classifier.py:127
↓ 5 callersClassInputExample
A single training/test example for simple sequence classification.
fastertransformer/examples/tensorflow/bert/bert-quantization/utils/create_glue_data.py:69
↓ 5 callersClassPredicatedTileIteratorParams
fastertransformer/3rdparty/cutlass/include/cutlass/epilogue/threadblock/predicated_tile_iterator_params.h:165
↓ 5 callersClassTestbedGrouped
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/test/gemm_grouped_testbed.py:41
↓ 5 callersClasscublasINT8MMWrapper
fastertransformer/src/fastertransformer/utils/cublasINT8MMWrapper.h:31
↓ 4 callersClassBertLayerNorm
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/modeling.py:333
↓ 4 callersClassBertTokenizer
Runs end-to-end tokenization: punctuation splitting + wordpiece
fastertransformer/examples/pytorch/bert/bert-quantization-sparsity/tokenization.py:95
↓ 4 callersClassCustomEncoder
fastertransformer/examples/pytorch/bert/utils/encoder.py:270
↓ 4 callersClassEncoderWeights
fastertransformer/examples/pytorch/bert/utils/encoder.py:25
↓ 4 callersClassFastLinearCombinationClamp
Applies a linear combination operator to an array of elements then clamps the output before converting to the output element type. D = a
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/epilogue.py:184
↓ 4 callersClassFeatureWriter
Writes InputFeature to TF example file.
fastertransformer/examples/tensorflow/bert/bert-quantization/utils/create_squad_data.py:472
↓ 4 callersClassGemmArguments
Argument wrapper for GEMM. It encodes problem information and user-provide tensors into the kernel's argument :param operation: the GEM
fastertransformer/3rdparty/cutlass/tools/library/scripts/pycutlass/src/pycutlass/gemm_operation.py:59
↓ 4 callersClassInputExample
A single training/test example for simple sequence classification.
fastertransformer/examples/tensorflow/xlnet/convertInput.py:70
next →1–100 of 4,979, ranked by callers