Functions — github.com/NVIDIA/FasterTransformer

CodeHub MCPcopy Create free account

hub / github.com/NVIDIA/FasterTransformer / functions

Functions4,228 in github.com/NVIDIA/FasterTransformer

⨍ Functions 4,228◇ Types & classes 1,417

↓ 1,438 callersMethodat

prevent converting int or size_t to string automatically

src/fastertransformer/utils/Tensor.h:353

↓ 827 callersMethodsize

src/fastertransformer/utils/Tensor.cc:164

↓ 640 callersMethodreMalloc

src/fastertransformer/utils/allocator.h:75

↓ 580 callersMethodfree

src/fastertransformer/utils/allocator.h:348

↓ 471 callersMethodinsert

src/fastertransformer/utils/Tensor.h:333

↓ 385 callersMethodcuda

examples/pytorch/gpt/utils/gpt.py:550

↓ 369 callersMethodGetInteger

3rdparty/INIReader.h:411

↓ 317 callersMethodget

src/fastertransformer/tensorrt_plugin/bert_fp8/bertFp8Plugin.h:78

↓ 262 callersMethodinfo

examples/pytorch/vit/ViT-quantization/quant_utils.py:30

↓ 197 callersMethoddata

examples/cpp/xlnet/cnpy.h:36

↓ 193 callersMethodend

src/fastertransformer/utils/Tensor.h:496

↓ 193 callersMethodget_tensor

src/fastertransformer/tf_op/BaseOp.h:68

↓ 189 callersMethodto

(self, device=None)

examples/pytorch/gpt/utils/gpt_decoder.py:954

↓ 171 callersMethodbegin

src/fastertransformer/utils/Tensor.h:491

↓ 115 callersMethodinfo

examples/pytorch/bert/bert-quantization-sparsity/quant_utils.py:31

↓ 113 callersMethodGemm

src/fastertransformer/utils/gemm.cc:23

↓ 111 callersMethodisExist

src/fastertransformer/utils/cublasAlgoMap.cc:125

↓ 110 callersFunctioninitialize

src/fastertransformer/utils/mpi_utils.cc:37

↓ 109 callersMethoddevice

examples/pytorch/gpt/utils/gpt_decoder.py:944

↓ 107 callersMethodclear

3rdparty/trt_fused_multihead_attention/fused_multihead_attention_v2.h:78

↓ 91 callersMethodload

(self, ckpt_path)

examples/pytorch/gpt/utils/gpt.py:539

↓ 87 callersMethodhalf

examples/pytorch/gpt/utils/gpt_decoder.py:964

↓ 86 callersMethodkeys

examples/pytorch/gptj/utils/reference_gptj.py:36

↓ 84 callersMethodfloat

examples/pytorch/gpt/utils/gpt_decoder.py:959

↓ 80 callersMethodGet

3rdparty/INIReader.h:394

↓ 74 callersFunctionfmtstr

src/fastertransformer/utils/string_utils.h:27

↓ 68 callersMethodGetFloat

3rdparty/INIReader.h:445

↓ 67 callersMethodlog

src/fastertransformer/utils/logger.h:47

↓ 64 callersFunctiongetSMVersion

src/fastertransformer/utils/cuda_utils.h:281

↓ 63 callersMethoddtype

examples/pytorch/gpt/utils/gpt_decoder.py:939

↓ 61 callersMethodsetFP32GemmConfig

src/fastertransformer/utils/cublasMMWrapper.cc:330

↓ 59 callersMethodrun

* \brief Forward a model and asynchronously check whether to stop. * * The device having the last rank of a pipeline parallel group checks a

examples/cpp/multi_gpu_gpt/multi_gpu_gpt_async_example.cc:338

↓ 52 callersMethodmax

src/fastertransformer/utils/Tensor.h:207

↓ 52 callersFunctionrotary_embedding_transform

src/fastertransformer/kernels/decoder_masked_multihead_attention_utils.h:1331

↓ 52 callersFunctiontorch2np

(tensor: torch.Tensor, np_data_type: typing.Optional[np.dtype] = None)

examples/pytorch/utils.py:28

↓ 51 callersMethodforward

(self, start_ids: torch.IntTensor, start_lengths: torch.IntTensor,

examples/pytorch/gpt/utils/gpt.py:580

↓ 50 callersMethodfrom_pretrained

Instantiate a PreTrainedBertModel from a pre-trained model file. Download and cache the pre-trained model file if needed.

examples/pytorch/bert/bert-quantization-sparsity/tokenization.py:140

↓ 49 callersFunctionas_GPU_tensor

src/fastertransformer/triton_backend/triton_utils.hpp:45

↓ 48 callersFunctionmove_tensor_H2D

src/fastertransformer/triton_backend/triton_utils.hpp:25

↓ 47 callersMethodsaveNpy

src/fastertransformer/utils/Tensor.cc:282

↓ 44 callersMethodsetFP16GemmConfig

src/fastertransformer/utils/cublasMMWrapper.cc:338

↓ 39 callersFunctionbarrier

examples/pytorch/bert/bert-quantization-sparsity/utils.py:40

↓ 39 callersMethodsetBF16GemmConfig

src/fastertransformer/utils/cublasMMWrapper.cc:347

↓ 37 callersFunctioncheckCublasStatus

tests/unittests/fp8_gemm_test/2022_03_21__fp8_stride_batch_example/worker.cpp:8

↓ 37 callersFunctioninvokeGeneralLayerNorm

src/fastertransformer/kernels/layernorm_kernels.h:138

↓ 36 callersFunctioncheckCudaStatus

tests/unittests/fp8_gemm_test/worker.cpp:98

↓ 36 callersMethodclose

examples/tensorflow/bert/bert-quantization/utils/create_squad_data.py:507

↓ 36 callersFunctiondeserialize_value

src/fastertransformer/tensorrt_plugin/swin/serialize.hpp:127

↓ 36 callersFunctionftNcclParamDestroy

src/fastertransformer/utils/nccl_utils.cc:299

↓ 36 callersFunctionserialize_value

src/fastertransformer/tensorrt_plugin/swin/serialize.hpp:121

↓ 35 callersMethodstridedBatchedGemm

src/fastertransformer/utils/gemm.cc:481

↓ 35 callersMethodwhat

src/fastertransformer/utils/gemm.h:601

↓ 34 callersMethodSpGemm

src/fastertransformer/utils/gemm.cc:715

↓ 34 callersFunctiongetActivationType

src/fastertransformer/utils/activation_types.h:34

↓ 34 callersMethodinfo

examples/pytorch/swin/Swin-Transformer-Quantization/quant_utils.py:30

↓ 34 callersFunctionrotary_embedding_coefficient

src/fastertransformer/kernels/decoder_masked_multihead_attention_utils.h:1325

↓ 32 callersMethodbfloat16

examples/pytorch/gpt/utils/gpt_decoder.py:969

↓ 32 callersFunctiondiffTime

src/fastertransformer/utils/cuda_utils.h:264

↓ 32 callersFunctionis_main_process

examples/pytorch/bert/bert-quantization-sparsity/utils.py:36

↓ 31 callersMethodGetBoolean

3rdparty/INIReader.h:468

↓ 30 callersFunctionsafe_transpose

(param: torch.nn.Parameter)

examples/pytorch/gpt/utils/huggingface_bloom_convert.py:126

↓ 29 callersFunctionftNcclRecv

src/fastertransformer/utils/nccl_utils.cc:111

↓ 29 callersFunctionftNcclSend

src/fastertransformer/utils/nccl_utils.cc:85

↓ 29 callersMethodupdate

(self, update_params: dict)

examples/pytorch/gpt/utils/gpt.py:742

↓ 28 callersMethodParseError

3rdparty/INIReader.h:382

↓ 28 callersFunctionnp2th

Possibly convert HWIO to OIHW.

examples/pytorch/vit/ViT-quantization/vit_int8.py:57

↓ 27 callersFunctiongetDevice

src/fastertransformer/utils/cuda_utils.h:317

↓ 27 callersFunctionpad_to_multiple_of_16

src/fastertransformer/kernels/moe_kernels.h:24

↓ 27 callersFunctionsum

src/fastertransformer/kernels/decoder_masked_multihead_attention_utils.h:1188

↓ 27 callersMethodupdate

(self, val, n=1)

examples/pytorch/vit/ViT-quantization/main.py:57

↓ 26 callersFunctioncheckCudaStatus

tests/unittests/fp8_gemm_test/2022_03_21__fp8_stride_batch_example/worker.cpp:15

↓ 26 callersFunctioncreate_initializer

(initializer_range=0.02, data_type=tf.float32)

examples/tensorflow/decoder/utils/common.py:247

↓ 26 callersFunctionftNcclBroadCast

src/fastertransformer/utils/nccl_utils.cc:136

↓ 25 callersMethod__init__

examples/pytorch/bert/bert-quantization-sparsity/modeling.py:534

↓ 25 callersFunctionbcast

src/fastertransformer/utils/mpi_utils.cc:114

↓ 25 callersFunctionbert_example

examples/pytorch/bert/bert_example.py:90

↓ 25 callersFunctioncheckCublasStatus

tests/unittests/fp8_gemm_test/worker.cpp:89

↓ 25 callersMethodcompressMatrix

src/fastertransformer/utils/cublasMMWrapper.cc:770

↓ 25 callersFunctionencoder_example

examples/pytorch/encoder/encoder_example.py:60

↓ 25 callersFunctionftNcclAllGather

src/fastertransformer/utils/nccl_utils.cc:70

↓ 24 callersMethodgetPtrWithOffset

src/fastertransformer/utils/Tensor.h:175

↓ 24 callersMethodisUseSparse

src/fastertransformer/utils/cublasAlgoMap.cc:209

↓ 24 callersMethodloadModel

src/fastertransformer/models/bert/BertWeight.cc:120

↓ 24 callersMethodto_host_vec

src/fastertransformer/utils/gpu_buf.h:65

↓ 23 callersMethodload

(module, prefix='')

examples/pytorch/bert/bert-quantization-sparsity/modeling.py:832

↓ 22 callersMethoddecode

examples/pytorch/gpt/utils/gpt_token_encoder.py:141

↓ 22 callersFunctiongetCublasDataType

src/fastertransformer/utils/gemm.cc:1024

↓ 21 callersMethodTensor

src/fastertransformer/utils/Tensor.cc:36

↓ 21 callersMethodsetGemmConfig

src/fastertransformer/utils/cublasMMWrapper.cc:356

↓ 20 callersFunctionftNcclInitialize

src/fastertransformer/utils/nccl_utils.cc:308

↓ 20 callersMethodsetStream

src/fastertransformer/models/t5/T5Encoder.cc:248

↓ 19 callersMethodall

src/fastertransformer/utils/Tensor.h:272

↓ 19 callersMethodforward

src/fastertransformer/th_op/vit/ViTOp.cc:99

↓ 19 callersMethodstart

(self, tag='__default')

examples/pytorch/gpt/bloom_lambada.py:98

↓ 18 callersFunctionftNcclAllReduceSum

src/fastertransformer/utils/nccl_utils.cc:56

↓ 18 callersFunctionftNcclGroupEnd

src/fastertransformer/utils/nccl_utils.cc:208

↓ 18 callersFunctionftNcclGroupStart

src/fastertransformer/utils/nccl_utils.cc:201

↓ 18 callersFunctionftNcclStreamSynchronize

src/fastertransformer/utils/nccl_utils.cc:215

↓ 18 callersMethodinsertIfValid

src/fastertransformer/utils/Tensor.h:340

↓ 18 callersFunctionprint_mem_usage

src/fastertransformer/utils/cuda_utils.h:271

next →1–100 of 4,228, ranked by callers