MCPcopy Create free account

hub / github.com/LitLeo/TensorRT_Tutorial / functions

Functions601 in github.com/LitLeo/TensorRT_Tutorial

↓ 1 callersFunctionmain
()
shenlan_homework_bert/builder.py:424
↓ 1 callersFunctionmake_pad_mask
(lengths: torch.Tensor, max_len: int = 0)
Hackathon2022/encoder_small.py:4
↓ 1 callersMethodmarkOutput
(self, x: trt.ITensor)
shenlan_homework_bert/trt_helper.py:77
↓ 1 callersFunctionmodel2onnx
(model, tokenizer, text)
shenlan_homework_bert/model2onnx.py:55
↓ 1 callersFunctionmodel_test
(model, tokenizer, text)
shenlan_homework_bert/model2onnx.py:18
↓ 1 callersFunctiononnx2trt
(onnxFile, plan_name)
Hackathon2022/decoder2trt.py:84
↓ 1 callersFunctiononnx2trt
(onnxFile, plan_name)
Hackathon2022/encoder2trt.py:300
↓ 1 callersFunctiononnx2trt
(onnxFile, plan_name)
shenlan_homework_bert/onnx2trt.py:18
↓ 1 callersFunctiononnx_to_trt_name
Converting variables in the onnx checkpoint to names corresponding to the naming convention used in the TF version, expected by the builder
shenlan_homework_bert/builder.py:202
↓ 1 callersFunctionremainder
Remainder of division. \param x first operand \param y second operand \return remainder of floating point division. template<typename T,typename U> ty
resource_for_billibilli/debug_plugin/plugin/half.h:2907
↓ 1 callersFunctionremquo
Remainder of division. \param x first operand \param y second operand \param quo address to store some bits of quotient at \return remainder of floati
resource_for_billibilli/debug_plugin/plugin/half.h:2931
↓ 1 callersFunctionround_half_up
Round half-precision number to nearest integer value using round-to-nearest-away-from-zero. \param value binary representation of half-precision value
resource_for_billibilli/debug_plugin/plugin/half.h:1348
↓ 1 callersFunctionrun
()
Hackathon2022/LayerNormPlugin/testLayerNormPlugin.py:59
↓ 1 callersFunctionrun
()
shenlan_homework_bert/基础款LayerNormPlugin/基础款LayerNormPlugin/testLayerNormPlugin.py:59
↓ 1 callersFunctionself_attention_layer
(network_helper, prefix, config, weights_dict, input_tensor, imask)
shenlan_homework_bert/builder.py:84
↓ 1 callersFunctionself_output_layer
(network_helper, prefix, config, weights_dict, hidden_states, input_tensor)
shenlan_homework_bert/builder.py:116
↓ 1 callersFunctionserialize_value
resource_for_billibilli/debug_plugin/plugin/serialize.hpp:111
↓ 1 callersMethodsetShouldLog
resource_for_billibilli/debug_plugin/plugin/logging.h:91
↓ 1 callersMethodset_layer_name
Tool function. Set the name of trt layer or plugin and print output shapes.
shenlan_homework_bert/trt_helper.py:25
↓ 1 callersFunctionset_tensor_name
(tensor, prefix, name)
shenlan_homework_bert/builder.py:56
↓ 1 callersFunctionsignbit
Check sign. \param arg number to check \retval true for negative number \retval false for positive number template<typename T> typename enable<bool,T>
resource_for_billibilli/debug_plugin/plugin/half.h:3863
↓ 1 callersFunctionsinh
Hyperbolic sine. \param arg function argument \return hyperbolic sine value of \a arg template<typename T> typename enable<expr,T>::type sinh(T arg) {
resource_for_billibilli/debug_plugin/plugin/half.h:3347
↓ 1 callersFunctiontan
Tangent function. \param arg function argument \return tangent value of \a arg template<typename T> typename enable<expr,T>::type tan(T arg) { return
resource_for_billibilli/debug_plugin/plugin/half.h:3268
↓ 1 callersFunctiontanh
Hyperbolic tangent. \param arg function argument \return hyperbolic tangent value of \a arg template<typename T> typename enable<expr,T>::type tanh(T
resource_for_billibilli/debug_plugin/plugin/half.h:3373
↓ 1 callersFunctiontest_case_data
(infer_helper, case_data_path)
shenlan_homework_bert/builder.py:396
↓ 1 callersFunctiontest_text
(infer_helper, BERT_PATH)
shenlan_homework_bert/builder.py:369
↓ 1 callersFunctiontgamma
Gamma function. \param arg function argument \return gamma function value of \a arg template<typename T> typename enable<expr,T>::type tgamma(T arg) {
resource_for_billibilli/debug_plugin/plugin/half.h:3468
↓ 1 callersFunctiontransformer_layer
(network_helper, prefix, config, weights_dict, input_tensor, imask)
shenlan_homework_bert/builder.py:138
↓ 1 callersFunctiontransformer_output_layer
(network_helper, config, weights_dict, input_tensor)
shenlan_homework_bert/builder.py:166
↓ 1 callersFunctiontrt_infer
(plan_name)
shenlan_homework_bert/onnx2trt.py:68
MethodCublasConfigHelper
resource_for_billibilli/debug_plugin/plugin/plugin_common.h:185
MethodDebugPlugin
It makes no sense to construct DebugPlugin without arguments.
resource_for_billibilli/debug_plugin/plugin/debug_plugin/debug_plugin.h:22
MethodDebugPluginDynamic
It doesn't make sense to make DebugPluginDynamic without arguments, so we delete default constructor.
resource_for_billibilli/debug_plugin/plugin/debug_plugin/debug_dynamic_plugin.h:21
MethodDims2
\brief Construct an empty Dims2 object.
resource_for_billibilli/debug_plugin/include/NvInfer.h:58
MethodDims3
\brief Construct an empty Dims3 object.
resource_for_billibilli/debug_plugin/include/NvInfer.h:157
MethodDims4
\brief Construct an empty Dims2 object.
resource_for_billibilli/debug_plugin/include/NvInfer.h:283
MethodDimsHW
\brief Construct an empty DimsHW object.
resource_for_billibilli/debug_plugin/include/NvInfer.h:88
FunctionEnumMax<ActivationType>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:126
FunctionEnumMax<BuilderFlag>
resource_for_billibilli/debug_plugin/include/NvInfer.h:6389
FunctionEnumMax<CalibrationAlgoType>
resource_for_billibilli/debug_plugin/include/NvInfer.h:5969
FunctionEnumMax<DataType>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:156
FunctionEnumMax<DeviceType>
resource_for_billibilli/debug_plugin/include/NvInferRuntime.h:714
FunctionEnumMax<DimensionOperation>
resource_for_billibilli/debug_plugin/include/NvInferRuntime.h:310
FunctionEnumMax<DimensionType>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:175
FunctionEnumMax<ElementWiseOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:2612
FunctionEnumMax<EngineCapability>
resource_for_billibilli/debug_plugin/include/NvInferRuntime.h:60
FunctionEnumMax<ErrorCode>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:1176
FunctionEnumMax<ErrorCode>
resource_for_billibilli/debug_plugin/include/NvOnnxParser.h:80
FunctionEnumMax<FillOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:4726
FunctionEnumMax<LayerType>
resource_for_billibilli/debug_plugin/include/NvInfer.h:471
FunctionEnumMax<LoopOutput>
resource_for_billibilli/debug_plugin/include/NvInfer.h:4488
FunctionEnumMax<MatrixOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:4143
FunctionEnumMax<NetworkDefinitionCreationFlag>
resource_for_billibilli/debug_plugin/include/NvInfer.h:6903
FunctionEnumMax<OptProfileSelector>
resource_for_billibilli/debug_plugin/include/NvInferRuntime.h:1007
FunctionEnumMax<PaddingMode>
resource_for_billibilli/debug_plugin/include/NvInfer.h:1169
FunctionEnumMax<PluginType>
resource_for_billibilli/debug_plugin/include/NvInferPluginUtils.h:53
FunctionEnumMax<PoolingType>
resource_for_billibilli/debug_plugin/include/NvInfer.h:1702
FunctionEnumMax<ProfilingVerbosity>
resource_for_billibilli/debug_plugin/include/NvInfer.h:6411
FunctionEnumMax<QuantizationFlag>
resource_for_billibilli/debug_plugin/include/NvInfer.h:6351
FunctionEnumMax<RNNDirection>
resource_for_billibilli/debug_plugin/include/NvInfer.h:2808
FunctionEnumMax<RNNGateType>
resource_for_billibilli/debug_plugin/include/NvInfer.h:3201
FunctionEnumMax<RNNInputMode>
resource_for_billibilli/debug_plugin/include/NvInfer.h:2836
FunctionEnumMax<RNNOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:2788
FunctionEnumMax<ReduceOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:3554
FunctionEnumMax<ResizeMode>
resource_for_billibilli/debug_plugin/include/NvInfer.h:4326
FunctionEnumMax<ScaleMode>
resource_for_billibilli/debug_plugin/include/NvInfer.h:2064
FunctionEnumMax<SliceMode>
resource_for_billibilli/debug_plugin/include/NvInfer.h:3876
FunctionEnumMax<TacticSource>
resource_for_billibilli/debug_plugin/include/NvInfer.h:6433
FunctionEnumMax<TensorFormat>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:321
FunctionEnumMax<TensorLocation>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:992
FunctionEnumMax<TopKOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:4050
FunctionEnumMax<TripLimit>
resource_for_billibilli/debug_plugin/include/NvInfer.h:4503
FunctionEnumMax<UnaryOperation>
resource_for_billibilli/debug_plugin/include/NvInfer.h:3492
FunctionEnumMax<WeightsRole>
resource_for_billibilli/debug_plugin/include/NvInferRuntime.h:696
MethodHALF_CONSTEXPR expr
Conversion constructor. \param f single-precision value to convert
resource_for_billibilli/debug_plugin/plugin/half.h:447
MethodLayerNormPlugin
Hackathon2022/LayerNormPlugin/LayerNormPlugin.h:48
MethodLayerNormPlugin
shenlan_homework_bert/基础款LayerNormPlugin/基础款LayerNormPlugin/LayerNormPlugin.h:46
MethodLayerNormPluginCreator
Hackathon2022/LayerNormPlugin/LayerNormPlugin.h:184
MethodLayerNormPluginCreator
shenlan_homework_bert/基础款LayerNormPlugin/基础款LayerNormPlugin/LayerNormPlugin.h:177
MethodLogStreamConsumer
\brief Creates a LogStreamConsumer which logs messages with level severity. Reportable severity determines if the messages are severe enough to be log
resource_for_billibilli/debug_plugin/plugin/logging.h:132
MethodLogStreamConsumerBase
resource_for_billibilli/debug_plugin/plugin/logging.h:109
MethodLogStreamConsumerBuffer
resource_for_billibilli/debug_plugin/plugin/logging.h:34
FunctionNvDimsOpMod
resource_for_billibilli/debug_plugin/plugin/plugin_common.h:172
MethodPluginRegistrar
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:1369
MethodSeverity>
resource_for_billibilli/debug_plugin/include/NvInferRuntimeCommon.h:1075
MethodSkipLayerNormV1Plugin
Hackathon2022/LayerNormPlugin/SkipLayerNormV1Plugin.h:32
MethodSkipLayerNormV1PluginCreator
Hackathon2022/LayerNormPlugin/SkipLayerNormV1Plugin.h:168
MethodSkipLayerNormV2Plugin
Hackathon2022/LayerNormPlugin/SkipLayerNormV2Plugin.h:32
MethodSkipLayerNormV2PluginCreator
Hackathon2022/LayerNormPlugin/SkipLayerNormV2Plugin.h:169
MethodTestAtom
resource_for_billibilli/debug_plugin/plugin/logging.h:264
Method__init__
(self, calibration_data_file, cache_file, batch_size)
Hackathon2022/calibrator.py:16
Method__init__
(self, network, plugin_registry, logger)
shenlan_homework_bert/trt_helper.py:18
Method__init__
(self, plan_name, trt_logger)
shenlan_homework_bert/trt_helper.py:227
Method__init__
(self, bert_config_path, use_fp16, use_int8, use_strict)
shenlan_homework_bert/builder.py:43
Method__init__
(self, data_txt, bert_path, cache_file, batch_size, max_seq_length, num_inputs)
shenlan_homework_bert/calibrator.py:32
Function_cudaGetErrorEnum
cublas&cudnn_int8_demo/cublasGemmEx/gemmInt8_rect.cpp:23
Methodacos
Arc cosine implementation. \param arg function argument \return function value stored in single-preicision
resource_for_billibilli/debug_plugin/plugin/half.h:1918
Methodacosh
Hyperbolic area cosine implementation. \param arg function argument \return function value stored in single-preicision
resource_for_billibilli/debug_plugin/plugin/half.h:1981
MethodaddConstant
(self, w, layer_name: Optional[str] = None)
shenlan_homework_bert/trt_helper.py:183
MethodaddLog
(self, x: trt.ITensor, layer_name=None, precision=None)
shenlan_homework_bert/trt_helper.py:148
← previousnext →101–200 of 601, ranked by callers