Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/Tencent/TurboTransformers
/ functions
Functions
392 in github.com/Tencent/TurboTransformers
⨍
Functions
392
◇
Types & classes
103
Method
from_npz
(file_name: str, layer_num: int)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:131
Method
from_npz
(file_name: str, layer_num: int, num_attention_heads: int)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:197
Method
from_npz
(file_name: str, num_hidden_layers: int, num_attention_heads: int)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:310
Method
from_npz
(file_name: str, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:363
Method
from_npz
(file_name: str, config, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:431
Method
from_npz
(file_name: str, config, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:584
Method
from_npz
(file_name: str, layer_num: int, num_attention_heads: int)
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:256
Method
from_npz
(file_name: str, config, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_gpt2.py:183
Method
from_npz
(file_name: str, layer_num: int)
turbo_transformers/python/turbo_transformers/layers/modeling_albert.py:178
Method
from_onmt
(multi_headed_attn: OnmtMultiHeadedAttention, is_trans_weight: bool = False)
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:153
Method
from_onmt
(position_wise_ffn: OnmtPositionwiseFeedForward, is_trans_weight: Optional[bool] = True)
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:290
Method
from_onmt
(transformer_decoder_layer: OnmtTransformerDecoderLayer)
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:420
Method
from_pretrained
(model_id_or_path: str, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:422
Method
from_pretrained
(model_id_or_path: str, device: Optional[torch.device] = None,
turbo_transformers/python/turbo_transformers/layers/modeling_gpt2.py:173
Method
from_torch
(bert_embedding: TorchBertEmbeddings)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:60
Method
from_torch
(intermediate: TorchBertIntermediate)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:90
Method
from_torch
(output: TorchBertOutput)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:122
Method
from_torch
(attention: TorchBertAttention)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:170
Method
from_torch
(encoder: TorchBertEncoder)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:303
Method
from_torch
(pooler: TorchBertPooler)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:355
Method
from_torch
(model: TorchBertModel, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:412
Method
from_torch
Args: model : a PyTorch Bert Model device : cpu or GPU backend : a string to indicates kernel provides
turbo_transformers/python/turbo_transformers/layers/modeling_bert.py:501
Method
from_torch
load an attn model from huggingface bert attention model.
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:182
Method
from_torch
(albert_embedding: TorchAlbertEmbeddings )
turbo_transformers/python/turbo_transformers/layers/modeling_albert.py:67
Method
from_torch
(attention: TorchAlbertAttention)
turbo_transformers/python/turbo_transformers/layers/modeling_albert.py:96
Method
from_torch
(torch_layer_group: TorchAlbertLayerGroup)
turbo_transformers/python/turbo_transformers/layers/modeling_albert.py:225
Method
from_torch
(torch_transformer_model: TorchAlbertTransformer)
turbo_transformers/python/turbo_transformers/layers/modeling_albert.py:293
Method
from_torch
(torch_model: TorchAlbertModel)
turbo_transformers/python/turbo_transformers/layers/modeling_albert.py:382
Method
from_torch
(intermediate)
turbo_transformers/python/turbo_transformers/layers/qmodeling_bert.py:25
Method
from_torch
(bert_output)
turbo_transformers/python/turbo_transformers/layers/qmodeling_bert.py:43
Method
from_torch
(encoder)
turbo_transformers/python/turbo_transformers/layers/qmodeling_bert.py:101
Method
from_torch
(model)
turbo_transformers/python/turbo_transformers/layers/qmodeling_bert.py:146
Method
from_torch
(model: TorchRobertaModel, device: Optional[torch.device] = None)
turbo_transformers/python/turbo_transformers/layers/modeling_roberta.py:142
Function
generate_onnx_model
(model_name: str, use_gpu: bool, filename: str,
benchmark/onnx_benchmark_helper.py:20
Method
init_state
Initialize decoder state.
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:528
Method
kMax>
turbo_transformers/layers/kernels/seq_pool.cpp:52
Method
kMean>
turbo_transformers/layers/kernels/seq_pool.cpp:36
Function
main
example/cpp/bert_model_example.cpp:124
Method
map_state
(self, fn)
turbo_transformers/python/turbo_transformers/layers/modeling_decoder.py:533
Method
mutableData
turbo_transformers/core/tensor.h:180
Method
operator()
example/cpp/bert_model.cpp:68
Method
operator()
do inference
example/cpp/bert_model.cpp:136
Method
operator()
turbo_transformers/core/tensor.h:36
Method
operator()
turbo_transformers/core/tensor.h:77
Method
operator()
turbo_transformers/core/tensor.h:311
Method
operator()
turbo_transformers/layers/albert_layer.cpp:31
Method
operator()
turbo_transformers/layers/bert_output.cpp:29
Method
operator()
turbo_transformers/layers/sequence_pool.cpp:21
Method
operator()
turbo_transformers/layers/bert_intermediate.cpp:32
Method
operator()
turbo_transformers/layers/multi_headed_attention.cpp:34
Method
operator()
turbo_transformers/layers/bert_embedding.cpp:25
Method
operator()
turbo_transformers/layers/addbias_layernorm.cpp:20
Method
operator()
turbo_transformers/layers/positionwise_ffn.cpp:31
Method
operator()
turbo_transformers/layers/bert_attention.cpp:29
Method
operator()
turbo_transformers/layers/bert_pooler.cpp:25
Method
operator()
turbo_transformers/layers/addbias_act.cpp:20
Method
operator()
turbo_transformers/layers/prepare_bert_masks.cpp:24
Function
operator+
turbo_transformers/core/half.h:66
Method
operator[]
turbo_transformers/core/tensor.h:247
Method
operator[]
turbo_transformers/loaders/npz_load.h:70
Function
pref_guard
(filename: str)
turbo_transformers/python/turbo_transformers/utils.py:34
Method
print_results
turbo_transformers/core/profiler.cpp:86
Function
run_model
(model, use_gpu, num_iter, batch_size, seq_len,
benchmark/benchmark_helper.py:17
Function
run_model
(model, use_cuda, num_iter=50, use_profile=False)
turbo_transformers/python/tests/test_helper.py:23
Function
run_variable_model
(model, use_gpu, num_iter, max_seq_len, min_seq_len, framework_name, num_threads, cfg)
benchmark/benchmark_helper.py:59
Method
setUp
(self)
turbo_transformers/python/tests/sequence_pool_test.py:36
Function
show_tensor
(T, info)
turbo_transformers/python/tests/test_helper.py:60
Method
start_profile
turbo_transformers/core/profiler.cpp:40
Method
test_Albert_attention
(self)
turbo_transformers/python/tests/albert_attention_test.py:96
Method
test_Roberta_model
(self)
turbo_transformers/python/tests/roberta_model_test.py:78
Method
test_bert_attention
(self)
turbo_transformers/python/tests/bert_attention_test.py:135
Method
test_bert_layer
(self)
turbo_transformers/python/tests/bert_layer_test.py:98
Method
test_bert_model
(self)
turbo_transformers/python/tests/bert_model_test.py:72
Method
test_bertout
(self)
turbo_transformers/python/tests/bert_output_test.py:80
Method
test_decoder
(self)
turbo_transformers/python/tests/decoder_transformer_decoder_layer_test.py:165
Method
test_dlpack
(self)
turbo_transformers/python/tests/tensor_conversion_test.py:36
Method
test_embedding
(self)
turbo_transformers/python/tests/bert_embedding_test.py:79
Method
test_embedding
(self)
turbo_transformers/python/tests/albert_embedding_test.py:92
Method
test_encoder
(self)
turbo_transformers/python/tests/bert_encoder_test.py:110
Method
test_gpt2_model
(self)
turbo_transformers/python/tests/gpt2_model_test.py:71
Method
test_intermediate
(self)
turbo_transformers/python/tests/bert_intermediate_test.py:87
Method
test_layer
(self)
turbo_transformers/python/tests/albert_layer_test.py:95
Method
test_layer
(self)
turbo_transformers/python/tests/albert_model_test.py:88
Method
test_multi_headed_attention
(self)
turbo_transformers/python/tests/decoder_multi_headed_attention_test.py:295
Method
test_pooler
(self)
turbo_transformers/python/tests/bert_pooler_test.py:86
Method
test_positionwise_feed_forward
(self)
turbo_transformers/python/tests/positionwise_ffn_test.py:99
Method
test_seq_pool
(self)
turbo_transformers/python/tests/sequence_pool_test.py:54
Function
vsTanh
turbo_transformers/core/blas_openblas.cpp:45
Method
what
turbo_transformers/core/enforce.cpp:24
Method
what
turbo_transformers/core/allocator.cpp:34
Method
~CUDADeviceContext
turbo_transformers/core/cuda_device_context.cpp:42
Method
~Half
turbo_transformers/core/half.h:42
← previous
301–392 of 392, ranked by callers