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

↓ 3 callersFunctionfloat_from_int8

src/fastertransformer/kernels/decoder_masked_multihead_attention/decoder_masked_multihead_attention_template.hpp:943

↓ 3 callersFunctionfloat_to_int8_rn_host

be consistent with FasterTransformer

src/fastertransformer/utils/convert_data_type.h:22

↓ 3 callersMethodflops

examples/pytorch/swin/Swin-Transformer-Quantization/models/swin_transformer_v2.py:498

↓ 3 callersMethodforward

src/fastertransformer/th_op/vit/ViTINT8Op.cc:104

↓ 3 callersMethodforward

src/fastertransformer/th_op/longformer/LongformerEncoderOp.cc:58

↓ 3 callersMethodforward

src/fastertransformer/th_op/swin/SwinOp.cc:128

↓ 3 callersMethodforward

src/fastertransformer/th_op/swin/SwinINT8Op.cc:131

↓ 3 callersMethodfrom_config

(cls, config: GptInitModelParameters, **kwargs)

examples/pytorch/gpt/utils/gpt_decoder.py:840

↓ 3 callersMethodfrom_json_file

Constructs a `BertConfig` from a json file of parameters.

examples/tensorflow/bert/bert-quantization/modeling.py:96

↓ 3 callersFunctionfuse3

examples/pytorch/vit/ViT-quantization/quant_utils.py:194

↓ 3 callersFunctionfuse3

examples/pytorch/bert/bert-quantization-sparsity/quant_utils.py:194

↓ 3 callersFunctionfuse3

examples/pytorch/swin/Swin-Transformer-Quantization/quant_utils.py:165

↓ 3 callersMethodfuse3

(self, qq, qk, qv)

examples/tensorflow/bert/bert-quantization/run_squad.py:286

↓ 3 callersMethodgating_softmax_test_helper

(self, dtype, rtol=1e-05, atol=1e-08)

tests/moe/th_moe_unit_tests.py:58

↓ 3 callersFunctiongetCublasComputeType

src/fastertransformer/utils/gemm.cc:1037

↓ 3 callersMethodgetFinishBuffer

src/fastertransformer/models/gptj/GptJ.cc:1237

↓ 3 callersFunctiongetGemmOpString

src/fastertransformer/utils/gemm.cc:1074

↓ 3 callersMethodgetSpAlgo

src/fastertransformer/utils/cublasAlgoMap.cc:196

↓ 3 callersFunctiongetUnfusedAttentionType

src/fastertransformer/layers/attention_layers/BaseAttentionLayer.h:128

↓ 3 callersMethodgetWeightsPtr

src/fastertransformer/models/longformer/LongformerEncoder.cc:134

↓ 3 callersMethodgetXMMAKernels

3rdparty/trt_fused_multihead_attention/fused_multihead_attention.h:410

↓ 3 callersFunctionget_args

examples/pytorch/gpt/utils/huggingface_bloom_convert.py:55

↓ 3 callersFunctionget_candidate_configs

src/fastertransformer/kernels/cutlass_kernels/cutlass_heuristic.cc:110

↓ 3 callersMethodget_dev_examples

Gets a collection of `InputExample`s for the dev set.

examples/pytorch/bert/bert-quantization-sparsity/processors/glue.py:61

↓ 3 callersFunctionget_name_prefix

examples/tensorflow/bert/bert-quantization/ft-tensorflow-quantization/ft_tensorflow_quantization/python/calib/calibrator.py:332

↓ 3 callersFunctionget_shape_list

(tensor, expected_rank=None, name=None)

examples/tensorflow/encoder/utils/encoder.py:135

↓ 3 callersFunctionget_tokenizer

(vocab_file=None, bpe_file=None)

examples/pytorch/gpt/utils/word_list.py:27

↓ 3 callersMethodget_train_examples

Gets a collection of `InputExample`s for the train set.

examples/pytorch/bert/bert-quantization-sparsity/processors/glue.py:57

↓ 3 callersFunctionget_world_size

examples/pytorch/gpt/utils/comm.py:19

↓ 3 callersFunctiongpt_generate_fn

examples/pytorch/gpt/multi_gpu_gpt_example.py:301

↓ 3 callersFunctionhandle_exceptions

(model_config: PretrainedConfig, param_name: str, param: torch.nn.

examples/pytorch/gpt/utils/huggingface_bloom_convert.py:205

↓ 3 callersFunctionigemm_config

src/fastertransformer/utils/gemm_test/encoder_igemm_func.cc:1013

↓ 3 callersFunctioninit_onmt_cache

(layer_num, memory_bank)

examples/pytorch/decoder/utils/decoder.py:48

↓ 3 callersFunctioninit_op_cache

(layer_num, batch_size, beam_width, max_seq_len, \ decoding_max_seq_len, head_num, size_per_

examples/pytorch/decoder/utils/decoder.py:30

↓ 3 callersMethodinit_state

Initialize decoder state.

examples/pytorch/decoding/utils/decoding.py:293

↓ 3 callersMethodinitialize

src/fastertransformer/models/vit/ViT.cc:31

↓ 3 callersFunctionisUnPaddedMHA

src/fastertransformer/layers/attention_layers/BaseAttentionLayer.h:118

↓ 3 callersFunctionlayer_norm

(input_tensor, name=None)

examples/tensorflow/encoder/utils/encoder.py:21

↓ 3 callersFunctionlayer_norm

Run layer normalization on the last dimension of the tensor.

examples/tensorflow/bert/bert-quantization/modeling.py:385

↓ 3 callersMethodload

(self, ckpt_path)

examples/pytorch/gptneox/utils/gptneox.py:237

↓ 3 callersFunctionload_state_dict

Load weights from model file `safetensors` or `pytorch binary` is supported # Args. file_path: model file path, ends with .bin or .

examples/pytorch/gpt/utils/huggingface_bloom_convert.py:308

↓ 3 callersFunctionload_the_npy_file

examples/cpp/xlnet/cnpy.cpp:201

↓ 3 callersMethodmap_state

examples/pytorch/decoding/utils/decoding.py:298

↓ 3 callersFunctionmetric_fn

Computes the loss and accuracy of the model.

examples/tensorflow/bert/bert-quantization/run_pretraining.py:276

↓ 3 callersFunctionnorm

Layer normalizes :obj:`inputs`.

examples/tensorflow/decoder/utils/decoder.py:28

↓ 3 callersFunctionparaName

examples/cpp/xlnet/xlnet_correctness_example.cc:133

↓ 3 callersFunctionparam2file

(pt_param, layer_id, save_dir, dest_key)

examples/pytorch/gptj/utils/gptj_ckpt_convert.py:90

↓ 3 callersFunctionparam2file

(pt_param, layer_id, save_dir, dest_key)

examples/pytorch/gptj/utils/huggingface_gptj_ckpt_convert.py:21

↓ 3 callersFunctionprint_abs_mean

(node, tensor, info = "")

examples/tensorflow/common_utils/common.py:126

↓ 3 callersMethodprocess_feature

Write a InputFeature to the TFRecordWriter as a tf.train.Example.

examples/tensorflow/bert/bert-quantization/utils/create_squad_data.py:481

↓ 3 callersFunctionquantize

(mat, act_range)

examples/pytorch/gpt/utils/huggingface_opt_convert.py:47

↓ 3 callersFunctionrandom_tensor

(shape, dtype, device, mean=0, std=1)

tests/gemm_dequantize/th_gemm_dequantize.py:4

↓ 3 callersFunctionreshape_from_matrix

Reshapes a rank 2 tensor back to its original rank >= 2 tensor.

examples/tensorflow/bert/tensorflow_bert/my_modeling.py:956

↓ 3 callersFunctionreshape_to_matrix

Reshapes a >= rank 2 tensor to a rank 2 tensor (i.e., a matrix).

examples/tensorflow/bert/utils/bert.py:299

↓ 3 callersFunctionreshape_to_matrix

Reshapes a >= rank 2 tensor to a rank 2 tensor (i.e., a matrix).

examples/tensorflow/bert/bert-quantization/modeling.py:1025

↓ 3 callersMethodrun_decode

(self, batch_size=4, beam_width=1, max_input_length=4

tests/unittests/th_op/test_th_decode_op.py:85

↓ 3 callersMethodset

src/fastertransformer/utils/gpu_buf.h:45

↓ 3 callersMethodsetMaxSeqLen

src/fastertransformer/models/gptneox/GptNeoXWeight.h:70

↓ 3 callersMethodset_hf_plugin_mode

(self, is_plugin)

examples/pytorch/longformer/model.py:80

↓ 3 callersFunctionset_tensor_common_0

3rdparty/fp8_qgmma_1x1/utils.h:73

↓ 3 callersFunctiontf_beamsearch_decoding

Run the decoding with beam search by TensorFlow. Args: memory_tensor: A tf.tensor with shape [batch_size * beam_width, max(memor

examples/tensorflow/decoder/utils/decoding.py:200

↓ 3 callersFunctiontf_encoder_opennmt

Run the bert transformer layer by TensorFlow. Args: input_tensor: A tf.Tensor with shape [batch_size, seq_len, hidden_dimension]

examples/tensorflow/encoder/utils/encoder.py:47

↓ 3 callersFunctiontf_sampling_decoding

Run the decoding with sampling by TensorFlow. Args: memory_tensor: A tf.tensor with shape [batch_size, max(memory_sequence_lengt

examples/tensorflow/decoder/utils/decoding.py:341

↓ 3 callersMethodto_cuda

examples/pytorch/vit/VisionTransformerINT8WeightLoader.py:97

↓ 3 callersMethodto_dict

Serializes this instance to a Python dictionary.

examples/tensorflow/bert/bert-quantization/modeling.py:102

↓ 3 callersMethodto_int8

(self, ths_path='../../../lib/libth_transformer.so')

examples/pytorch/vit/VisionTransformerINT8WeightLoader.py:149

↓ 3 callersFunctionto_numpy_dtype

(maybe_str_dtype: Union[str, np.dtype])

examples/pytorch/gpt/utils/gpt_decoder.py:32

↓ 3 callersFunctiontrain_loop

(args, model, optimizer, step, num_steps)

examples/pytorch/bert/bert-quantization-sparsity/apex_sparsity/test/checkpointing_test_reference.py:35

↓ 3 callersMethodtranspose_for_scores

examples/pytorch/vit/ViT-quantization/vit_int8.py:170

↓ 3 callersFunctiontrt_getS

examples/cpp/swin/functions.h:43

↓ 3 callersFunctionvalidate

(config, data_loader, model)

examples/pytorch/swin/Swin-Transformer-Quantization/main.py:300

↓ 3 callersMethodwarn

examples/pytorch/bert/bert-quantization-sparsity/quant_utils.py:33

↓ 3 callersFunctionwhitespace_tokenize

Runs basic whitespace cleaning and splitting on a piece of text.

examples/tensorflow/bert/bert-quantization/tokenization.py:138

↓ 2 callersMethodCopyWeightsFromHostBuffers

src/fastertransformer/models/vit/ViTWeight.h:203

↓ 2 callersMethodDynamicDecodeOp

src/fastertransformer/th_op/common/DynamicDecodeOp.cc:211

↓ 2 callersMethodGetSerializeSize

src/fastertransformer/models/vit_int8/ViTINT8Weight.h:155

↓ 2 callersMethodGetWeightCount

src/fastertransformer/models/vit/ViTWeight.h:236

↓ 2 callersFunctionLtIgemmCustomFind

src/fastertransformer/utils/gemm_test/encoder_igemm_func.cc:281

↓ 2 callersMethodSwinTransformerINT8Class

src/fastertransformer/th_op/swin/SwinINT8Op.cc:22

↓ 2 callersMethod__init__

examples/pytorch/bert/utils/modeling_bert.py:26

↓ 2 callersMethod__init__

(self, model_type=None, data_type=None, decoder_ths_path=None, decoding_ths_path=None)

examples/pytorch/decoding/utils/decoding.py:416

↓ 2 callersMethod__init__

(self, num_heads: int, size_per_head: int, num_layers: int,

examples/pytorch/gpt/utils/gpt_decoder.py:714

↓ 2 callersFunction__nv_fp8_e4m3

tests/unittests/fp8_gemm_test/2022_03_21__fp8_stride_batch_example/include/cuda_fp8.hpp:706

↓ 2 callersFunction_append_to_models

(model, adapter, rank_weights, is_merge: bool)

examples/pytorch/t5/utils/nemo_t5_ckpt_convert.py:904

↓ 2 callersFunction_compute_softmax

Compute softmax probability over raw logits.

examples/pytorch/bert/bert-quantization-sparsity/run_squad.py:695

↓ 2 callersFunction_config_update_adapter

examples/pytorch/t5/utils/nemo_t5_ckpt_convert.py:1127

↓ 2 callersMethod_create_examples

Creates examples for the training and dev sets.

examples/tensorflow/xlnet/convertInput.py:216

↓ 2 callersMethod_create_examples

(self, data_dir)

examples/tensorflow/xlnet/convertInput.py:240

↓ 2 callersMethod_create_examples

Creates examples for the training and dev sets.

examples/pytorch/bert/bert-quantization-sparsity/processors/glue.py:103

↓ 2 callersMethod_create_examples

Creates examples for the training and dev sets.

examples/pytorch/bert/bert-quantization-sparsity/processors/glue.py:142

↓ 2 callersMethod_create_examples

Creates examples for the training and dev sets.

examples/pytorch/bert/bert-quantization-sparsity/processors/glue.py:181

↓ 2 callersMethod_create_examples

Creates examples for the training and dev sets.

examples/pytorch/bert/bert-quantization-sparsity/processors/glue.py:215

↓ 2 callersMethod_create_position_encoding_table

(self, max_seq_len, input_dim, dtype)

examples/tensorflow/decoder/utils/position.py:62

↓ 2 callersMethod_decode_and_generate

( self, decoder_in, memory_bank, batch, src_vocabs

examples/pytorch/decoding/utils/translator.py:509

↓ 2 callersMethod_encode

Creates position encodings. Args: positions: The positions to encode of shape :math:`[B, ...]`. depth: The encoding depth

examples/tensorflow/bert/utils/position.py:52

↓ 2 callersMethod_encode

Creates position encodings. Args: positions: The positions to encode of shape :math:`[B, ...]`. depth: The encoding depth

examples/tensorflow/decoder/utils/position.py:52

↓ 2 callersMethod_encode

Creates position encodings. Args: positions: The positions to encode of shape :math:`[B, ...]`. depth: The encoding depth

examples/tensorflow/common_utils/position.py:52

↓ 2 callersFunction_extract_pp_weights

(model, pp_idx: int)

examples/pytorch/t5/utils/nemo_t5_ckpt_convert.py:873

↓ 2 callersFunction_fused_gemm_dq

tests/gemm_dequantize/th_gemm_dequantize.cc:78

↓ 2 callersFunction_get_best_indexes

Get the n-best logits from a list.

examples/tensorflow/bert/bert-quantization/run_squad.py:912

← previous next →401–500 of 4,228, ranked by callers