►Ndmlc | |
►Nserializer | |
CHandler< DLDataType > | |
CHandler< DLDevice > | |
►Ntvm | Runtime implementation for LibTorch/TorchScript |
►Narith | Namespace of arithmetic analysis |
CAnalyzer | Analyzer that contains bunch of sub-analyzers |
CCanonicalSimplifier | Canonical-form based simplifier |
CConstIntBound | Reference class to ConstIntBoundNode |
CConstIntBoundAnalyzer | Analyzer to get constant integer bound over expression |
CConstIntBoundNode | Constant integer up and lower bound(inclusive). Useful for value bound analysis |
CConstraintContext | Constraint context |
CIntConstraints | Managed reference to IntConstraintsNode |
CIntConstraintsNode | Represent integer constrains including (integer) variables, their ranges and the relations between them (either equations or inequalities) |
CIntConstraintsTransform | Managed reference to IntConstraintsTransformNode |
CIntConstraintsTransformNode | We can have different set of variables to represent the same constraints. For example, the following two systems are equivalent, {a + b = 0 | a >= 0, b >= 0} and {m - n = 0 | m >= 0, n <= 0} This data structure represents the transformation between two equivalent linear systems. In the above example, src : {a + b = 0 | a >= 0, b >= 0} dst : {m - n = 0 | m >= 0, n <= 0} src_to_dst : {a -> m, b -> -n} dst_to_src : {m -> a, n -> -b} |
CIntGroupBounds | Managed reference to IntGroupBoundsNode |
CIntGroupBoundsNode | Represent integer grouped bounds which are classified into lower bounds (inclusive), upper bounds (inclusive) and equalities. It also contains coefficient as a multiplier for the bounds, i.e., coef * var >= lower coef * var == equal coef * var <= upper |
CIntSet | Managed reference to IntSetNode |
CIntSetAnalyzer | Integer set analyzer |
CIntSetNode | Base class of all Integer set containers. represent a set of integers in one dimension |
CIterMapExpr | Managed reference to IterMapExprNode |
CIterMapExprNode | Base class of all iter map expressions |
CIterMapResult | Managed reference to IterMapResultNode |
CIterMapResultNode | Result of DetectIterMap |
CIterMark | Managed reference to IterMarkExprNode |
CIterMarkNode | Mark the source as an iterator in [0, extent) |
CIterSplitExpr | Managed reference to IterSplitExprNode |
CIterSplitExprNode | Split of an iterator |
CIterSumExpr | Managed reference to IterSumExprNode |
CIterSumExprNode | Fuse multiple iterators by summing them with scaling |
CModularSet | Reference of ModularSetNode |
CModularSetAnalyzer | Analyzer to get modular information over expression |
CModularSetNode | Range of a linear integer function. Use to do specify the possible index values |
CRewriteSimplifier | Rewrite-rule based simplifier |
CTransitiveComparisonAnalyzer | Using previously specified knowns, compare the expressions provided |
►Nauto_scheduler | |
CAccessAnalyzer | Managed reference to AccessAnalyzerNode |
CAccessAnalyzerNode | Static analyzer for a ComputeDAG |
CAnnotationStep | Managed reference to AnnotationStepNode |
CAnnotationStepNode | Annotation step that corresponds to vectorize, parallel, unroll and thread binding. (i.e. te::Stage::vectorize, te::Stage::parallel, te::Stage::vectorize, te::Stage::bind) |
CAttachMap | Managed reference to AttachMapNode |
►CAttachMapNode | Stores the compute_at relation between stages This stores a bi-directional mapping from stages and iter: |
CIterKeyHash | |
CBuildResult | Managed reference to BuildResultNode |
CBuildResultNode | Store the result of a build |
CCacheReadStep | Managed reference to CacheReadStepNode |
CCacheReadStepNode | Cache read step that corresponds to te::Schedule::cache_read |
CCacheWriteStep | Managed reference to CacheWriteStepNode |
CCacheWriteStepNode | Cache write step that corresponds to te::Schedule::cache_write |
CComputeAtStep | Managed reference to ComputeAtStepNode |
CComputeAtStepNode | Compute at step that corresponds to te::Stage::compute_at |
CComputeDAG | Managed reference to ComputeDAGNode |
CComputeDAGNode | The auto-scheduler's computational graph and related program analyses |
CComputeInlineStep | Managed reference to ComputeInlineStepNode |
CComputeInlineStepNode | Compute inline step that corresponds to te::Stage::compute_inline |
CComputeRootStep | Managed reference to ComputeRootStepNode |
CComputeRootStepNode | Compute root step that corresponds to te::Stage::compute_root |
CCostModel | Managed reference to CostModelNode |
CCostModelNode | The base class for cost model |
CFollowFusedSplitStep | Managed reference to FollowFusedSplitStepNode |
CFollowFusedSplitStepNode | Similar to FollowSplitStep, but uses split factors from multiple steps |
CFollowSplitStep | Managed reference to FollowSplitStepNode |
CFollowSplitStepNode | Similar to SplitStepNode, but uses split factors from another step (i.e. Follow another split step) |
CFuseStep | Managed reference to FuseStepNode |
CFuseStepNode | Fuse step that corresponds to te::Stage::fuse |
CHardwareParams | Managed reference to HardwareParamsNode |
CHardwareParamsNode | The parameters of target hardware used to guide the SearchPolicy |
CIterator | Managed reference to IteratorNode |
CIteratorNode | An iterator of a for-loop Similar to tvm::IterVar in include/tvm/tir/expr.h |
CLocalBuilder | Managed reference to LocalBuilderNode |
CLocalBuilderNode | LocalBuilder use local CPU cores to build programs in parallel |
CLocalRunner | Managed reference to LocalRunnerNode |
CLocalRunnerNode | LocalRunner that uses local CPU/GPU to measure the time cost of programs |
CMeasureCallback | Managed reference to MeasureCallbackNode |
CMeasureCallbackNode | Bass class of measurement callbacks |
CMeasureInput | Managed reference to MeasureInputNode |
CMeasureInputNode | Store the input of a measurement |
CMeasureResult | Managed reference to MeasureResultNode |
CMeasureResultNode | Store the results of a measurement |
CPragmaStep | Managed reference to PragmaStepNode |
CPragmaStepNode | Pragma step that corresponds to te::Stage::pragma |
CPreloadMeasuredStates | Managed reference to PreloadMeasuredStatesNode |
CPreloadMeasuredStatesNode | Preload measured states from a log file. This can resume the state of the search policy |
CProgramBuilder | Managed reference to ProgramBuilderNode |
CProgramBuilderNode | ProgramBuilder that builds the programs |
CProgramMeasurer | Managed reference to ProgramMeasurerNode |
CProgramMeasurerNode | Measurer that measures the time costs of tvm programs This class combines ProgramBuilder and ProgramRunner, and provides a simpler API |
CProgramRunner | Managed reference to ProgramRunnerNode |
CProgramRunnerNode | ProgramRunner that runs the built programs and measure the time cost |
CPythonBasedMeasureCallback | Managed reference to PythonBasedMeasureCallbackNode |
CPythonBasedMeasureCallbackNode | A wrapper for measure callback defined by python code This class will call functions defined in the python |
CPythonBasedModel | Managed reference to PythonBasedModelNode |
CPythonBasedModelNode | A wrapper for cost model defined by python code This class will call functions defined in the python |
CRandomModel | Managed reference to RandomModelNode |
CRandomModelNode | The cost model returning random value for all predictions |
CRecordReader | Managed reference to RecordReaderNode |
CRecordReaderNode | Log reader to load step logs from a file |
CRecordToFile | Managed reference to RecordToFileNode |
CRecordToFileNode | Callback for logging the input and results of measurements to file |
CReorderStep | Managed reference to ReorderStepNode |
CReorderStepNode | Reorder step that corresponds to te::Stage::reorder |
CRfactorStep | Managed reference to RfactorStepNode |
CRfactorStepNode | Reduction factor step that corresponds to te::Schedule::rfactor |
CRPCRunner | Managed reference to RPCRunnerNode |
CRPCRunnerNode | RPCRunner that uses RPC call to measures the time cost of programs on remote devices. Or sometime we may need to use RPC even in local running to insulate the thread environment. (e.g. running CUDA programs) |
CSearchCallback | Managed reference to SearchCallbackNode |
CSearchCallbackNode | Callback function to be called by the search process. This interface allows to do extra initializations before schedule search or extra check during/after the schedule search |
CSearchPolicy | Managed reference to SearchPolicyNode |
CSearchPolicyKey | Attribute keys of ops used for SearchPolicy |
CSearchPolicyNode | The base class of search policies |
CSearchTask | Managed reference to SearchTaskNode |
CSearchTaskNode | The computation information and hardware parameters for a specific schedule search task |
CSplitStep | Managed reference to SplitStepNode |
CSplitStepNode | Split step that corresponds to te::Stage::split with additional support of multiple-level of factors |
CStage | Managed reference to StageNode |
CStageAttributes | Stage-level attributes |
CStageNode | A op stage in the compute declaration. Similar to te::Stage in include/tvm/te/schedule.h |
CState | Managed reference to StateNode |
CStateNode | A state in the search process. It consists of the current loop structure and a list of transformation steps used to construct it. Each State corresponds to a specific schedule for its ComputeDAG |
CStep | Managed reference to StepNode |
CStepNode | The base class of transformation steps. Each step has its corresponding tvm.te schedule primitives |
CStorageAlignStep | Managed reference to StorageAlignStepNode |
CStorageAlignStepNode | Storage align step that corresponds to te::Stage::storage_align |
CTuningOptions | Managed reference to TuningOptionsNode |
CTuningOptionsNode | Tuning and measurement options |
►Ndetail | |
CAttrDocEntry | |
CAttrDocVisitor | |
CAttrExistVisitor | |
CAttrInitEntry | |
CAttrInitVisitor | |
CAttrNonDefaultVisitor | |
CAttrNopEntry | |
CAttrNormalVisitor | |
CAttrsSEqualVisitor | |
CAttrsSHashVisitor | |
CAttrTriggerNonDefaultEntry | |
CImplSEqualReduce | |
CImplSEqualReduce< T, true > | |
CImplSHashReduce | |
CImplSHashReduce< T, true > | |
CImplVisitAttrs | |
CImplVisitAttrs< T, true > | |
Cis_specialized | |
Cis_specialized< Container< Args... >, Container > | |
CReflectionTrait | |
CSelectSEqualReduce | |
CSelectSEqualReduce< T, TraitName, false > | |
CSelectSHashReduce | |
CSelectSHashReduce< T, TraitName, false > | |
CSelectVisitAttrs | |
CSelectVisitAttrs< T, TraitName, false > | |
CTypeName | Helper struct to get the type name known to tvm |
CTypeName< bool > | |
CTypeName< DataType > | |
CTypeName< double > | |
CTypeName< int > | |
CTypeName< int64_t > | |
CTypeName< uint64_t > | |
CTypeName< void * > | |
CValueTypeInfoMaker | |
►Ninstrument | |
CPassInstrument | Managed reference class for PassInstrumentNode |
CPassInstrumentNode | PassInstrumentNode forms an instrument implementation. It provides API for users to register callbacks at different instrumentation points |
►Nmeta_schedule | |
CArgInfo | Managed reference to ArgInfoNode |
CArgInfoNode | The argument information |
CBuilder | Managed reference to BuilderNode |
CBuilderInput | Managed reference to BuilderInputNode |
CBuilderInputNode | The builder's input, containing an IRModule and the target |
CBuilderNode | The abstract builder interface |
CBuilderResult | Managed reference to BuilderResultNode |
CBuilderResultNode | The builder's output, containing the artifact path or error message if any |
CCostModel | Managed reference to CostModelNode |
CCostModelNode | Cost model |
CDatabase | Managed reference to DatabaseNode |
CDatabaseNode | |
CExtractedTask | Managed reference to ExtractedTaskNode |
CExtractedTaskNode | A tuning task extracted from the high-level IR |
CFeatureExtractor | Managed reference to FeatureExtractorNode |
CFeatureExtractorNode | Extractor for features from measure candidates for use in cost model |
CMeasureCallback | Managed reference to MeasureCallbackNode |
CMeasureCallbackNode | Rules to apply after measure results is available |
CMeasureCandidate | Managed reference to MeasureCandidateNode |
CMeasureCandidateNode | The schedule (with input shapes) to be measured |
CMutator | Managed reference to MutatorNode |
CMutatorNode | Mutator is designed to mutate the trace to explore the design space |
CPostproc | Managed reference to PostprocNode |
CPostprocNode | Rules to apply a postprocessor to a schedule |
CProfiler | Managed reference to ProfilerNode |
CProfilerNode | A generic profiler |
CPyBuilderNode | An abstract builder with customized build method on the python-side |
CPyCostModelNode | The cost model with customized methods on the python-side |
CPyDatabaseNode | The database with customized methods on the python-side |
CPyFeatureExtractorNode | The feature extractor with customized methods on the python-side |
CPyMeasureCallbackNode | The measure callback with customized methods on the python-side |
CPyMutatorNode | The mutator with customized methods on the python-side |
CPyPostprocNode | The postprocessor with customized methods on the python-side |
CPyRunnerNode | An abstract runner with customized build method on the python-side |
CPyScheduleRuleNode | The schedule rule with customized methods on the python-side |
CPySearchStrategyNode | The python side customizable class for measure candidate generation |
CPySpaceGeneratorNode | The design space generator with customized methods on the python-side |
CPyTaskSchedulerNode | The task scheduler with customized methods on the python-side |
CRunner | Managed reference to RunnerNode |
CRunnerFuture | Managed reference to RunnerFutureNode |
CRunnerFutureNode | A class to asynchronously fetch runner's output |
CRunnerInput | Managed reference to RunnerInputNode |
CRunnerInputNode | Runner's input containing path of artifact, type of device and argument info |
CRunnerNode | The abstract runner interface |
CRunnerResult | Managed reference to RunnerResultNode |
CRunnerResultNode | Runner's output containing measurement result of MeasureCandidate or error msg if any |
CScheduleRule | Managed reference to ScheduleRuleNode |
CScheduleRuleNode | Rules to modify a block in a schedule |
CScopedTimer | |
CSearchStrategy | Managed reference to SearchStrategyNode |
CSearchStrategyNode | The search strategy for measure candidates generation |
CSpaceGenerator | Managed reference to SpaceGeneratorNode |
CSpaceGeneratorNode | The abstract class for design space generation |
CTaskRecord | Managed reference to TaskRecordNode |
CTaskRecordNode | |
CTaskScheduler | Managed reference to TaskSchedulerNode |
CTaskSchedulerNode | The abstract interface of task schedulers |
CTensorInfo | Managed reference to TensorInfoNode |
CTensorInfoNode | The tensor argument information |
CTuneContext | Managed reference to TuneContextNode |
CTuneContextNode | The auto tuning context |
CTuningRecord | The managed reference of TuningRecordNode |
CTuningRecordNode | The class of tuning records |
CWorkload | Managed reference to WorkloadNode |
CWorkloadEqual | The equality check for Workload |
CWorkloadHash | The hash method for Workload |
CWorkloadNode | A workload, i.e. an IRModule and its structural hash |
►Nrelay | Relay: a high level functional IR for TVM |
►Nqnn | |
CBroadcastAttrs | Attribute for broadcast operator |
CDequantizeAttrs | Attribute for dequantize operator |
CQuantizeAttrs | Attribute for quantize operator |
CRequantizeAttrs | Attribute for requantize operator |
CSimulatedQuantizeAttrs | |
CAdaptivePool1DAttrs | Attributes for 1d adaptive pool operator |
CAdaptivePool2DAttrs | Attributes for 2d adaptive pool operator |
CAdaptivePool3DAttrs | Attributes for 3d adaptive pool operator |
CAffineGridAttrs | Attributes used in image affine_grid operator |
CAllClassNonMaximumSuppressionAttrs | Attributes used in all_class_non_maximum_suppression operator |
CAllocStorageAttrs | Options for allocating storage |
CAllocTensorAttrs | Options for allocating tensors |
CAltPattern | A pattern which matches either of two patterns |
CAltPatternNode | Pattern for Alternate Expressions |
CArangeAttrs | Attributes used in arange operators |
CArgReduceAttrs | Attributes for Reduce operators which reduce by finding a single element. E.g. argmin |
CArgsortAttrs | Attributes used in argsort operators |
CAttrPattern | A pattern which matches attributes in another pattern |
CAttrPatternNode | Pattern for Attributes |
CAutoSchedulerLayoutTransformAttrs | Attributes for AutoSchedulerLayoutTransform operator |
CAvgPool1DAttrs | Attributes for 1D avg pool operator |
CAvgPool2DAttrs | Attributes for avg pool operator |
CAvgPool3DAttrs | Attributes for 3D avg pool operator |
CBatchMatmulAttrs | Attributes for batch matmul operator |
CBatchNormAttrs | Attributes used in batch_norm operator |
CBatchToSpaceNDAttrs | Attributes used in BatchToSpaceND operator |
CBiasAddAttrs | Add a 1D Tensor to an axis of a data |
CBinaryConv2DAttrs | Attribues used in bitserial convolution operators |
CBinaryDenseAttrs | |
CBitPackAttrs | Attributes used in bitpack operators |
CCall | |
CCallLoweredAttrs | Metadata for calls to TIR functions, useful for program analysis crossing Relay and TIR |
CCallNode | Call container |
CCallPattern | |
CCallPatternNode | CallPattern container |
CCastAttrs | Data type cast |
CCastHintAttrs | Annotate an expression to be cast into specific data type |
CClause | |
CClauseNode | Clause container node |
CClipAttrs | Attributes for Clip operator |
CCompileError | Custom Error class to be thrown during compilation |
CCompilerAttrs | Options for the operators used to annotate a compiler |
CConcatenateAttrs | Attributes used in concatenate operators |
CConstant | |
CConstantNode | Constant tensor type |
CConstantPattern | |
CConstantPatternNode | Container for Constant |
CConstructorValue | |
CConstructorValueObj | |
CConv1DAttrs | Attributes used in 1D convolution operators |
CConv1DTransposeAttrs | Attributes used in 1D transposed convolution operator |
CConv2DAttrs | Attributes used in convolution operators |
CConv2DTransposeAttrs | Attributes used in transposed convolution operator |
CConv2DWinogradAttrs | Attributes used in convolution operators with winograd algorithm |
CConv2DWinogradNNPACKWeightTransformAttrs | Attributes used in winograd weight transformation operators |
CConv3DAttrs | Attributes used in convolution operators |
CConv3DTransposeAttrs | Attributes used in transposed convolution operator |
CConv3DWinogradAttrs | Attributes used in 3d winograd convolution operators |
CConvGemmWeightTransformAttrs | Attributes used in gemm weight transformation operators |
CConvWinogradWeightTransformAttrs | Attributes used in winograd weight transformation operators |
CCorrelationAttrs | Attributes used in correlation operators |
CCropAndResizeAttrs | Attributes used in image crop_and_resize operator |
CDataTypePattern | A pattern which matches a type in another pattern |
CDataTypePatternNode | Pattern for Types |
CDebugAttrs | Options for the debug operators |
CDeformableConv2DAttrs | Attributes for DeformableConv2D operator |
CDenseAttrs | Attributes for dense operator |
CDensePackAttrs | Attributes for dense_pack operator |
CDeviceCopyAttrs | Options for the device copy operators |
CDFPattern | Managed reference to dataflow patterns |
CDFPatternCallback | Managed reference to dataflow pattern callbacks |
CDFPatternCallbackNode | Base type of all dataflow pattern callbacks |
CDFPatternFunctor | A dynamical functor that dispatches on in the first DFPattern argument |
CDFPatternFunctor< R(const DFPattern &n, Args...)> | |
CDFPatternNode | Base type of all dataflow patterns |
CDFPatternVisitor | A simple visitor wrapper around DFPatternFunctor. Recursively visit the content |
CDFTAttrs | Attributes used in DFT operator |
CDilateAttrs | Attributes used in dilate operator |
CDilation2DAttrs | Attributes used in dilation operators |
CDominatorPattern | A pattern which matches a variable length dominator path |
CDominatorPatternNode | Dominated Graph Pattern Pattern for fuzzy subgraphs where all outputs of the parent are used finally by the child, and every operation between the parent and the child matches the path |
CDropoutAttrs | Attributes used in dropout operator |
CDynExpandDimsAttrs | Attributes used in dynamic expand_dims operators |
CEinsumAttrs | Attributes used in einsum operator |
CErrorBuilder | A wrapper around std::stringstream to build error. include/tvm/ir/type.h Can be consumed by CompileError to construct an error |
CErrorReporter | An abstraction around how errors are stored and reported. Designed to be opaque to users, so we can support a robust and simpler error reporting mode, as well as a more complex mode |
CExecutor | Managed reference class to ExecutorNode |
CExecutorNode | Executor information |
CExecutorRegEntry | Helper structure to register Executors |
CExpandDimsAttrs | Attributes used in expand_dims operators |
CExprFunctor | A dynamical functor that dispatches on in the first Expr argument. You can use this as a more powerful Visitor, since it allows you to define function signatures of Visit Function |
CExprFunctor< R(const Expr &n, Args...)> | |
CExprMutator | A wrapper around ExprFunctor which functionally updates the AST |
CExprPattern | A pattern which matches a literal expression |
CExprPatternNode | Pattern for Relay Expression |
CExprRewriter | A non-iterating Expression Rewriter |
CExprVisitor | A simple visitor wrapper around ExprFunctor. Recursively visit the content |
CFeatureSet | A finite set of Feature |
CFIFOBufferAttrs | Attributes for FIFO buffer operator |
CFixedPointMultiplyAttrs | Attributes for FixedPointMultiply operator |
CFixedPointMultiplyPerAxisAttrs | Attributes for per channel/per axes FixedPointMultiply operator |
CFunction | Managed reference to FunctionNode |
CFunctionNode | Relay Function container |
CFunctionPattern | Managed reference to FunctionNode |
CFunctionPatternNode | Relay Function container |
CGatherAttrs | |
CGatherNDAttrs | |
CGetValidCountsAttrs | Attributes used in get_valid_counts operator |
CGlobalPool2DAttrs | Attributes for global pool operator |
CGridSampleAttrs | Attributes used in image grid_sample operator |
CGroupNormAttrs | Attributes used in group_norm operator |
CId | |
CIdNode | The unique identifier of variables |
CIf | |
CIfNode | Container of If |
CIfPattern | |
CIfPatternNode | |
CInitOpAttrs | Attributes that specify a tensor |
CInstanceNormAttrs | Attributes used in instance_norm operator |
CInterpreterClosure | |
CInterpreterClosureObj | The container type of Closures used by the interpreter |
CL2NormalizeAttrs | Attributes for L2Normalize operator |
CLayerNormAttrs | Attributes used in layer_norm operator |
CLayoutTransformAttrs | Attributes for LayoutTransform operator |
CLeakyReluAttrs | Attributes for leaky relu operator |
CLet | |
CLetNode | A binding of a sub-network |
CLetPattern | Let binding that binds a local var |
CLetPatternNode | A binding of a sub-network |
CLRNAttrs | Attributes for LRN operator |
CMatch | |
CMatchNode | Match container node |
CMatmulAttrs | Attributes for matmul operator |
CMatrixSetDiagAttrs | Attributes used in matrix_set_diag operator |
CMaxPool1DAttrs | Attributes for 1D max pool operator |
CMaxPool2DAttrs | Attributes for max pool operator |
CMaxPool3DAttrs | Attributes for 3D max pool operator |
CMeshgridAttrs | Attributes used in meshgrid operators |
CMetaScheduleLayoutTransformAttrs | Attributes for MetaScheduleLayoutTransform operator |
CMirrorPadAttrs | Attributes used for the MirrorPadding operator |
CMixedModeMutator | Non-recursive DFS Graph Traversal for Custom Rewriting Passes |
CMixedModeVisitor | A wrapper around ExprVisitor which traverses the Dataflow Normal AST |
CMultiBoxPriorAttrs | Attributes used in multibox_prior operators |
CMultiBoxTransformLocAttrs | |
CMultinomialAttrs | |
CNdarraySizeAttrs | Attributes for ndarray_size operator |
CNLLLossAttrs | Attributes used in NLLLoss operator |
CNonMaximumSuppressionAttrs | Attributes used in non_maximum_suppression operator |
CNormalAttrs | |
COnDeviceAttrs | Attributes for the "on_device" annotation (ie operator) |
COneHotAttrs | Attributes used in one-hot operator |
COpImplementation | Operator implementation class |
COpImplementationNode | Operator implementation that includes compute and schedule function |
COpSpecialization | Operator specialization class |
COpSpecializationNode | Specialized implementations for operators under certain conditions |
COpStrategy | Operator strategy class |
COpStrategyNode | Operator strategy to choose implementation |
CPadAttrs | Attributes used for the padding operator |
CPattern | Pattern is the base type for an ADT match pattern in Relay |
CPatternConstructor | |
CPatternConstructorNode | PatternVar container node |
CPatternFunctor | A dynamical functor on ADT patterns that dispatches on its first argument. You can use this as a more powerful visitor, since it allows you to define the types of further arguments to VisitPattern |
CPatternFunctor< R(const Pattern &n, Args...)> | |
CPatternMutator | A wrapper around ExprFunctor which functionally updates the AST |
CPatternNode | Base type for declaring relay pattern |
CPatternTuple | |
CPatternTupleNode | PatternVar container node |
CPatternVar | |
CPatternVarNode | PatternVar container node |
CPatternVisitor | A simple visitor wrapper around PatternFunctor |
CPatternWildcard | |
CPatternWildcardNode | PatternWildcard container node |
CPReluAttrs | Attributes for prelu operator |
CProposalAttrs | Attributes used in proposal operators |
CRecClosure | |
CRecClosureObj | The container type of RecClosure |
CReduceAttrs | Attributes for Reduce operators |
CRefCreate | |
CRefCreateNode | |
CRefRead | |
CRefReadNode | |
CRefValue | |
CRefValueObj | |
CRefWrite | |
CRefWriteNode | |
CRelayNode | This is the base node container of all relay structures |
CRepeatAttrs | Attributes used in repeat operators |
CReshapeAttrs | Attributes used in reshape operators |
CReshapeLikeAttrs | Attributes used in MXNet-style reshape_like operators |
CReshapeTensorAttrs | Attributes for VM reshape_tensor operator |
CResize1DAttrs | Attributes used in image resize1d operator |
CResize2DAttrs | Attributes used in image resize2d operator |
CResize3DAttrs | Attributes used in image resize3d operator |
CReverseAttrs | Attributes used in reverse operators |
CReverseSequenceAttrs | Attributes used in reverse_sequence operators |
CROIAlignAttrs | Attributes used in roi_align operators |
CROIPoolAttrs | Attributes used in roi_pool operators |
CRuntime | Managed reference class to RuntimeNode |
CRuntimeNode | Runtime information |
CRuntimeRegEntry | Helper structure to register Runtimes |
CScanopAttrs | Attributes used in cumsum and cumprod operator |
CScatterElementsAttrs | |
CScatterNDAttrs | |
CSearchSortedAttrs | |
CSequenceMaskAttrs | |
CShapeFuncAttrs | Options for the shape function operator |
CShapeOfAttrs | Attributes for ShapeOf operator |
CShapePattern | A pattern which matches a type in another pattern |
CShapePatternNode | Pattern for Shapes |
CSliceLikeAttrs | |
CSlidingWindowAttrs | Attributes used for the sliding_window operator |
CSoftmaxAttrs | Attributes used in softmax operators |
CSpaceToBatchNDAttrs | Attributes used in SpaceToBatchND operator |
CSparseConv2DAttrs | Attributes for sparse_dense operator |
CSparseDenseAttrs | Attributes for sparse_dense operator |
CSparseToDenseAttrs | Attributes used in sparse_to_dense operator |
CSparseTransposeAttrs | Attributes for sparse_transpose operator |
CSplitAttrs | |
CSqueezeAttrs | Attributes used in squeeze operators |
CStackAttrs | Attributes used in stack operators |
CStftAttrs | Attributes used in stft operator |
CStridedSliceAttrs | Attributes for StridedSlice operator |
CSubPixelAttrs | Attributes used in subpixel operators |
CTakeAttrs | |
CTempExpr | |
CTempExprNode | Base class of the temporary expression |
CThreefryGenerateAttrs | |
CTileAttrs | Attributes used in tile operators |
CTopKAttrs | |
CTransposeAttrs | Attributes used in transpose operators |
CTriluAttrs | |
CTuple | |
CTupleGetItem | |
CTupleGetItemNode | |
CTupleGetItemPattern | |
CTupleGetItemPatternNode | |
CTupleNode | Tuple container |
CTuplePattern | |
CTuplePatternNode | Tuple container |
CTypePattern | A pattern which matches a type in another pattern |
CTypePatternNode | Pattern for Types |
CUniformAttrs | |
CUniqueAttrs | Attributes used in unique operator |
CUpSampling3DAttrs | Attributes for upsampling3d operator |
CUpSamplingAttrs | Attributes for upsampling operator |
Cv_info | A struct to keep info of traversed expr in ExpandDataflow function |
CVar | |
CVarianceAttrs | |
CVarNode | Container for Var |
CVarPattern | |
CVarPatternNode | Container for Var |
CWildcardPattern | A pattern which matches anything |
CWildcardPatternNode | Wildcard Pattern |
CYoloReorgAttrs | Attributes used in yolo reorg operators |
►Nruntime | |
►Nmetadata | |
CArrayAccessor | A span-like class which permits access to Array fields with complex elements. These array fields should be accessed from C++ using the Metadata wrapper classes. This class lazily instantiates those wrappers as they are accessed |
CArrayAccessor< const char *, ::tvm::runtime::String > | A specialization of ArrayAccessor for String. This class is needed because the String constructor signature is different from the typical Metadata subclass |
CArrayIterator | An iterator implementation that lazily instantiates the C++ wrapping Metadata class |
CConstantInfoMetadata | |
CConstantInfoMetadataNode | |
CMetadata | |
CMetadataArray | Reference class for MetadataArray |
CMetadataArrayNode | Container for arrays in the metadata |
CMetadataBase | Reference class for the common MetadataBaseNode class |
CMetadataBaseNode | Common base class for all Metadata |
CMetadataNode | |
CTensorInfo | |
CTensorInfoNode | |
►Nmicro_rpc | |
CFrameBuffer | |
CFramer | |
CPacketFieldSizeBytes | |
CSession | CRT communication session management class. Assumes the following properties provided by the underlying transport: |
CSessionHeader | |
CUnframer | |
CWriteStream | |
►Nprofiling | |
CCallFrame | |
CCountNode | |
CDeviceWrapper | Wrapper for Device |
CDeviceWrapperNode | Wrapper for Device because Device is not passable across the PackedFunc interface |
CDurationNode | |
CMetricCollector | Wrapper for MetricCollectorNode |
CMetricCollectorNode | Interface for user defined profiling metric collection |
CPercentNode | |
CProfiler | |
CRatioNode | |
CReport | |
CReportNode | Data collected from a profiling run. Includes per-call metrics and per-device metrics |
►Nthreading | |
CThreadGroup | A platform-agnostic abstraction for managing a collection of thread pool threads |
►Nvm | |
CAllocator | |
CBuffer | |
CExecutable | The executable emitted by the VM compiler |
CInstruction | A single virtual machine instruction |
CMemoryManager | |
CStorage | Reference to storage |
CStorageObj | An object representing a storage allocation |
CVirtualMachine | The virtual machine |
CVMClosure | Reference to closure |
CVMClosureObj | An object representing a vm closure |
CVMFrame | A representation of a stack frame |
CVMFunction | A representation of a Relay function in the VM |
CADT | Reference to algebraic data type objects |
CADTObj | An object representing a structure or enumeration |
►CArray | Array, container representing a contiguous sequence of ObjectRefs |
CValueConverter | |
CArrayNode | Array node content in array |
CClosure | Reference to closure |
CClosureObj | An object representing a closure. This object is used by both the Relay VM and interpreter |
CDataType | Runtime primitive data type |
CDenseMapNode | A specialization of hash map that implements the idea of array-based hash map. Another reference implementation can be found [1] |
CDeviceAPI | TVM Runtime Device API, abstracts the device specific interface for memory management |
CInplaceArrayBase | Base template for classes with array like memory layout |
Cis_valid_iterator | Helper struct for type-checking |
Cis_valid_iterator< Optional< T >, IterType > | |
CIterAdapter | Iterator adapter that adapts TIter to return another type |
►CMap | Map container of NodeRef->NodeRef in DSL graph. Map implements copy on write semantics, which means map is mutable but copy will happen when array is referenced in more than two places |
Citerator | Iterator of the hash map |
►CMapNode | Shared content of all specializations of hash map |
Citerator | |
CModule | Module container of TVM |
CModuleNode | Base container of module |
►CNDArray | Managed NDArray. The array is backed by reference counted blocks |
CContainer | Object container class that backs NDArray |
CContainerBase | The container base structure contains all the fields except for the Object header |
CNullOptType | Helper to represent nullptr for optional |
CObjAllocatorBase | Base class of object allocators that implements make. Use curiously recurring template pattern |
CObject | Base class of all object containers |
CObjectEqual | String-aware ObjectRef hash functor |
CObjectHash | String-aware ObjectRef equal functor |
CObjectPtr | A custom smart pointer for Object |
CObjectPtrEqual | ObjectRef equal functor |
CObjectPtrHash | ObjectRef hash functor |
CObjectRef | Base class of all object reference |
CObjectTypeChecker | Type traits for runtime type check during FFI conversion |
CObjectTypeChecker< Array< T > > | |
CObjectTypeChecker< Map< K, V > > | |
COptional | Optional container that to represent to a Nullable variant of T |
CPackedFunc | Packed function is a type-erased function. The arguments are passed by packed format |
►CPackedFuncObj | Object container class that backs PackedFunc |
CExtractor | Internal struct for extracting the callable method from callable type |
CPackedFuncSubObj | Derived object class for constructing PackedFuncObj |
CPackedFuncValueConverter | Type trait to specify special value conversion rules from TVMArgValue and TVMRetValue |
CPackedFuncValueConverter< Optional< T > > | |
CPackedFuncValueConverter< PrimExpr > | |
CPackedFuncValueConverter< tvm::Bool > | |
CPackedFuncValueConverter< tvm::Integer > | |
CPackedFuncValueConverter<::tvm::runtime::String > | |
CRegistry | Registry for global function |
CReverseIterAdapter | Iterator adapter that adapts TIter to return another type |
CShapeTuple | Reference to shape tuple objects |
►CShapeTupleObj | An object representing a shape tuple |
CFromStd | An object representing shape tuple moved from std::vector |
CSignaturePrinter | |
►CSimpleObjAllocator | |
CArrayHandler | |
CHandler | |
CSmallMapNode | A specialization of small-sized hash map |
CString | Reference to string objects |
►CStringObj | An object representing string. It's POD type |
CFromStd | An object representing string moved from std::string |
CTimer | Timer for a specific device |
CTimerNode | Base class for all implementations |
CTVMArgs | Arguments into TVM functions |
CTVMArgsSetter | |
CTVMArgValue | A single argument value to PackedFunc. Containing both type_code and TVMValue |
CTVMMovableArgValue_ | Internal auxiliary struct for TypedPackedFunc to indicate a movable argument |
CTVMMovableArgValueWithContext_ | Internal auxiliary struct for TypedPackedFunc to indicate a movable argument with additional context information (function name and argument index) for better error reporting |
CTVMPODValue_ | Internal base class to handle conversion to POD values |
CTVMRetValue | Return Value container, Unlike TVMArgValue, which only holds reference and do not delete the underlying container during destruction |
CTypedPackedFunc | Please refer to TypedPackedFunc<R(Args..)> |
CTypedPackedFunc< R(Args...)> | A PackedFunc wrapper to provide typed function signature. It is backed by a PackedFunc internally |
CTypeIndex | Namespace for the list of type index |
►Nscript | |
►Nir_builder | |
►Ndetails | |
CNamer | |
►Nir | |
CIRModuleFrame | Managed reference to IRModuleFrameNode |
CIRModuleFrameNode | A frame that represents the IRModule frame with functions and global variables |
►Ntir | |
CAllocateConstFrame | Managed reference to AllocateConstFrameNode |
CAllocateConstFrameNode | A frame represents the allocate constant |
CAllocateFrame | Managed reference to AllocateFrameNode |
CAllocateFrameNode | A frame represents the allocate |
CAssertFrame | Managed reference to AssertFrameNode |
CAssertFrameNode | A frame that represents the assert statement. Proceeds if the condition is true, otherwise aborts with the message |
CAttrFrame | Managed reference to AttrFrameNode |
CAttrFrameNode | A frame that represents attribute node |
CBlockFrame | Managed reference to BlockFrameNode |
CBlockFrameNode | A frame that represents the block |
CBlockInitFrame | Managed reference to BlockInitFrameNode |
CBlockInitFrameNode | A frame that represents the block initialization statment |
CDeclBufferFrame | |
CDeclBufferFrameNode | |
CElseFrame | Managed reference to ElseFrameNode |
CElseFrameNode | A frame that represents else |
CForFrame | Managed reference to ForFrameNode |
CForFrameNode | A frame that represents the for loop |
CIfFrame | Managed reference to IfFrameNode |
CIfFrameNode | A frame that represents if statement |
CLaunchThreadFrame | Managed reference to LaunchThreadFrameNode |
CLaunchThreadFrameNode | The LaunchThreadFrameNode |
CLetFrame | Managed reference to LetFrameNode |
CLetFrameNode | A frame represents the let binding expression, which binds a var |
CPrimFuncFrame | Managed reference to PrimFuncFrameNode |
CPrimFuncFrameNode | A frame that represents the PrimFunc containing TIR statements |
CRealizeFrame | Managed reference to RealizeFrameNode |
CRealizeFrameNode | A frame that represents realization |
CThenFrame | Managed reference to ThenFrameNode |
CThenFrameNode | A frame that represents then |
CTIRFrame | Managed reference to TIRFrameNode |
CTIRFrameNode | A base frame that represents the TIR fame with body of statements |
CWhileFrame | Managed reference to WhileFrameNode |
CWhileFrameNode | A frame that represents while loop |
CIRBuilder | Managed reference to an IRBuilderNode |
CIRBuilderFrame | Managed reference to an IRBuilderFrameNode |
CIRBuilderFrameNode | |
CIRBuilderNode | A dialect-agnostic IRBuilder that constructs any IR of TVM. An idiomatic use of this class is to put this inside the RAII with-scope, call dialect-specific methods accordingly. Upon exiting the scope |
►Nprinter | |
CAssertDoc | Reference type of AssertDocNode |
CAssertDocNode | Doc that represents assert statement |
CAssignDoc | Reference type of AssignDocNode |
CAssignDocNode | Doc that represents assign statement |
CAttrAccessDoc | Reference type of AttrAccessDocNode |
CAttrAccessDocNode | Doc that represents attribute access on another expression |
CCallDoc | Reference type of CallDocNode |
CCallDocNode | Doc that represents function call |
CClassDoc | Reference type of ClassDocNode |
CClassDocNode | Doc that represents class definition |
CCommentDoc | Reference type of CommentDocNode |
CCommentDocNode | Doc that represents comment |
CDictDoc | Reference type of DictDocNode |
CDictDocNode | Doc that represents dictionary literal |
CDoc | Reference type of DocNode |
CDocNode | The base class of all Doc |
CDocStringDoc | Reference type of DocStringDocNode |
CDocStringDocNode | Doc that represents docstring |
CExprDoc | Reference type of ExprDocNode |
CExprDocNode | The base class of expression doc |
CExprStmtDoc | Reference type of ExprStmtDocNode |
CExprStmtDocNode | Doc that represents an expression as statement |
CForDoc | Reference type of ForDocNode |
CForDocNode | Doc that represents for statement |
CFrame | Reference type of FrameNode |
CFrameNode | |
CFunctionDoc | Reference type of FunctionDocNode |
CFunctionDocNode | Doc that represents function definition |
CIdDoc | Reference type of IdDocNode |
CIdDocNode | Doc that represents identifier |
CIfDoc | Reference type of IfDocNode |
CIfDocNode | Doc that represent if-then-else statement |
CIndexDoc | Reference type of IndexDocNode |
CIndexDocNode | Doc that represents index access on another expression |
CIRDocsifier | Reference type of IRDocsifierNode |
CIRDocsifierFunctor | Dynamic dispatch functor based on ObjectPath |
►CIRDocsifierNode | IRDocsifier is the top-level interface in the IR->Doc process |
CVariableInfo | Information about a variable, including its optional name and its doc creator |
CLambdaDoc | Reference type of LambdaDocNode |
CLambdaDocNode | Doc that represents anonymous function |
CListDoc | Reference type of ListDocNode |
CListDocNode | Doc that represents list literal |
CLiteralDoc | Reference type of LiteralDocNode |
CLiteralDocNode | Doc that represents literal value |
COperationDoc | Reference type of OperationDocNode |
COperationDocNode | Doc that represents operation |
CReturnDoc | Reference type of ReturnDocNode |
CReturnDocNode | Doc that represents return statement |
CScopeDoc | Reference type of ScopeDocNode |
CScopeDocNode | Doc that represents special scopes |
CSliceDoc | Reference type of SliceDocNode |
CSliceDocNode | Doc that represents slice in Index expression |
CStmtBlockDoc | Reference type of StmtBlockDocNode |
CStmtBlockDocNode | The container doc that holds a list of StmtDoc |
CStmtDoc | Reference type of StmtDocNode |
CStmtDocNode | The base class of statement doc |
CTupleDoc | Reference type of TupleDocNode |
CTupleDocNode | Doc that represents tuple literal |
CWhileDoc | Reference type of WhileDocNode |
CWhileDocNode | Doc that represents while statement |
►Nsupport | |
CLinearCongruentialEngine | This linear congruential engine is a drop-in replacement for std::minstd_rand. It strictly corresponds to std::minstd_rand and is designed to be platform-independent |
►CSpan | A partial implementation of the C++20 std::span |
Citerator_base | |
►Nte | Tensor expression language DSL |
CBaseComputeOpNode | A Compute op that compute a tensor on certain domain. This is the base class for ComputeOp (operating on a scalar at a time) and TensorComputeOp (operating on a TensorSlice at a time) |
CComputeOp | Managed reference to ComputeOpNode |
CComputeOpNode | A Compute op that compute a tensor on certain domain |
CExternOp | Managed reference to ExternOpNode |
CExternOpNode | External computation that cannot be splitted |
CFuse | Managed reference to FuseNode |
CFuseNode | Fuse two domains into one domain |
CHybridOp | Managed reference to HybridOpNode |
CHybridOpNode | A computation operator that generated by hybrid script |
CIterVarAttr | Additional scheduable attributes about IterVar |
CIterVarAttrNode | Node container for IterVar attr |
CIterVarRelation | The schedule relation between IterVars can be Split, Fuse |
CIterVarRelationNode | Base node of iteration var |
COperation | Operation that produces tensors |
COperationNode | Base class of all operation nodes |
CPlaceholderOp | Managed reference to PlaceholderOpNode |
CPlaceholderOpNode | A placeholder op represents an input placeholder |
CRebase | Managed reference to RebaseNode |
CRebaseNode | Rebase the iteration to make min to be 0. This is useful to normalize the Schedule to make every leaf variable's min to be 0 |
CScanOp | Managed reference to ScanOpNode |
CScanOpNode | Symbolic scan |
CSchedule | Global schedule container For operations and all the operations they depend on. The schedule per Operation is named as stage |
CScheduleContext | Context helper to collect debug information of Schedule |
CScheduleNode | Node container for schedule |
CSingleton | Managed reference to SingletonNode |
CSingletonNode | Singleton iterator [0, 1) |
CSpecializedCondition | Specialized condition to enable op specialization |
CSpecializedConditionNode | Container for specialization conditions |
CSplit | Managed reference to SplitNode |
CSplitNode | Split the parent domain into product of outer and iter |
CStage | Stage, contains scheduling for a stage of computation |
CStageNode | Stage |
►CTensor | Tensor structure representing a possible input, or intermediate computation result |
CSlice | Data structure to represent a slice that fixes first k coordinates. This is used to enable syntax sugar of Tensor[x][y][z] to get the element |
CTensorComputeOp | Managed reference to TensorComputeOpNode |
CTensorComputeOpNode | A TenorCompute op that compute a tensor with an tensor intrinsic |
CTensorDom | Temporary data structure to store union of bounds of each axis of Tensor |
CTensorIntrin | Managed reference to TensorIntrinNode |
CTensorIntrinCall | Managed reference to TensorIntrinCallNode |
CTensorIntrinCallNode | |
CTensorIntrinNode | Node to represent a Tensor intrinsic operator |
CTensorNode | Node to represent a tensor |
CTransform | |
CTransformNode | Transform iterator according to some arbitrary expression |
►Ntir | |
►Nusmp | |
►Nalgo | |
CGreedyBase | This is the base class for Greedy Algorithms where the sorting is specialized in the extended classes based on the greedy criteria |
CAllocatedPoolInfo | |
CAllocatedPoolInfoNode | This object contains information post-allocation for PoolInfo objects |
CBufferInfo | |
CBufferInfoAnalysis | |
CBufferInfoAnalysisNode | This is a composite node that is produced by extract_buffer_info analysis pass that contains useful global information that could be useful for memory planning algorithms |
CBufferInfoNode | Describes an abstract memory buffer that will get allocated inside a pool. The actual memory buffer in represented by PoolAllocationNode after static memory planning |
CPoolAllocation | |
CPoolAllocationNode | The pool allocation produced after the USMP algorithm |
CAdd | Managed reference to AddNode |
CAddNode |
|
CAllocate | Managed reference to AllocateNode |
CAllocateConst | Managed reference to AllocateConstNode |
CAllocateConstNode | Allocate a buffer that can be used in body |
CAllocateNode | Allocate a buffer that can be used in body |
CAnd | Managed reference to AndNode |
CAndNode | && b |
CAny | Managed reference to AnyNode |
CAnyNode | Any shape |
CAssertStmt | Managed reference to AssertStmtNode |
CAssertStmtNode | Assert condition, if an error occurs, return the error message |
CAttrStmt | Managed reference to AttrStmtNode |
CAttrStmtNode | Define certain auxiliary attribute for the body to be a symbolic value. This provide auxiliary information for IR passes that transforms body |
CBijectiveLayout | Bijective function mapping for data layout transformation. Given two Layout, BijectiveLayout build and store the mapping rules, provides API to transform N-dimention tensor from the source indices (i0, i1, .., im) to the destination indices (j0, j1, .., jm) |
CBijectiveLayoutNode | |
CBinaryOpNode | Base template to implement binary ops |
CBlock | Managed reference to BlockNode |
CBlockInfo | The information about a TensorIR block, it contains two categories of information 1) Info on the block scope rooted at a specific block, including dependency tracking, flags indicating if the scope is a stage pipeline, etc. 2) Info on the block itself, including if the block has a quasi-affine binding, if the regions it reads are completely covered by their producers, etc |
CBlockNode | A block is a basic schedule unit in TIR |
CBlockRealize | Managed reference to BlockRealizeNode |
CBlockRealizeNode | A block realization node represents execution of the block at the binding values |
CBlockRV | Managed reference to BlockRVNode |
CBlockRVNode | A random variable that evaluates to a TensorIR block |
CBlockScope | Managed reference to BlockScopeNode |
CBlockScopeNode | An object with 1-to-1 correspondence with each block reference in the sref tree. This data structure is used to track the producer-consumer dependencies between blocks. For example even leaf nodes have a scope node, even though they have no dependencies |
CBroadcast | Managed reference to BroadcastNode |
CBroadcastNode | Create a vector where all the elements are value |
CBuffer | Buffer is a symbolic n-darray structure. It is a composition of primitive symbolic types, used to specify the memory layout of the Tensor used in program input |
CBufferLoad | Managed reference to BufferLoadNode |
CBufferLoadNode | Load value from the high dimension buffer |
CBufferNode | Node to represent a buffer |
CBufferRealize | Managed reference to BufferRealizeNode |
CBufferRealizeNode | Annotate the region where the buffer need to be read and write in the body. We only need to allocate the space for the corresponding region |
CBufferRegion | Managed reference to BufferRegionNode |
CBufferRegionNode | Representing the region of multi-dimensional buffer access |
CBufferStore | Managed reference to BufferStoreNode |
CBufferStoreNode | Store value to the high dimension buffer |
CCall | Managed reference to CallNode |
CCallNode | Call node |
CCast | Managed reference to CastNode |
CCastNode | Cast value from one data type to another |
CCmpOpNode | Base template to implement comparison ops |
CCommReducer | Managed reference to CommReducerNode |
CCommReducerNode | A commutative reducer node to represent a commutative binary operator with identity element |
CDataProducer | Managed reference to DataProducerNode |
CDataProducerNode | Base node for data producers |
CDataTypeLegalizer | Legalize the data types of expressions to make sure they are consistent with other parts of the program |
CDeclBuffer | Managed reference to DeclBufferNode |
CDeclBufferNode | Declare a buffer that can be used in the body |
CDependency | Managed reference to DependencyNode |
CDependencyNode | A tuple (src, dst, kind) representing certain types of dependency. For example, (A, B, kRAW) means block B depends on block A, and the dependency kind is read-after-write, which means block B reads the result written by block A |
CDiv | Managed reference to DivNode |
CDivNode | / b in the C semnatics |
CEQ | Managed reference to EQNode |
CEQNode | == b |
CEvaluate | Managed reference to EvaluateNode |
CEvaluateNode | Evaluates an expression. This is mostly used for putting a Call node into Stmt |
CExprDeepEqual | Compare two expressions recursively and check if they are equal to each other without var remapping |
CExprFunctor | A dynamical functor that dispatches on in the first Expr argument. You can use this as a more powerful Visitor, since it allows you to define function signatures of Visit Function |
CExprFunctor< R(const PrimExpr &n, Args...)> | |
CExprMutator | ExprMutator that mutates expressions |
CExprVisitor | ExprVisitor |
CFloorDiv | Managed reference to FloorDivNode |
CFloorDivNode | Floor division, floor(a/b) |
CFloorMod | Managed reference to FloorModNode |
CFloorModNode | The remainder of the floordiv |
CFor | Managed reference to ForNode |
CForNode | A for loop, with poissible type annotations |
CGE | Managed reference to GENode |
CGENode | >= b |
CGT | Managed reference to GTNode |
CGTNode | > b |
CIfThenElse | Managed reference to IfThenElseNode |
CIfThenElseNode | IfThenElse statment |
CIndexDataTypeNormalizer | Normalize the data types of buffer shapes and indices to the same data type |
CIndexDataTypeRewriter | Data type rewriter for buffer indices |
CIndexMap | |
CIndexMapNode | Defines a mapping between two representations of indices into a buffer |
CInstruction | Managed reference to InstructionNode |
CInstructionKind | Managed reference to InstructionKindNode |
CInstructionKindNode | Kind of an instruction, e.g. Split, Reorder, etc. Besides the name, every kind of instruction has its own properties, including: 1) A boolean indicating if the instruction is pure, i.e. change nothing in the schedule state 2) A functor that applies the instruction to a TensorIR schedule 3) A functor that converts the instruction to a statement in python syntax 4) A functor that serialize its attributes to JSON 5) A functor that deserialize its attributes from JSON |
CInstructionKindRegEntry | An entry in the registry of InstructionKind |
CInstructionNode | Schedule instructions each corresponds to a schedule primitive |
CIterVar | Iteration Variable, represents an iteration over an integer interval |
CIterVarNode | An iteration variable representing an iteration over a one dimensional interval |
CLayout | Managed reference to LayoutNode |
CLayoutAxis | |
CLayoutNode | Layout is to describe how data is organized within an N-dimention tensor. It is composed of upper cases, lower cases and numbers, where upper case indicates a primal axis and the corresponding lower case with factor size indicates the subordinate axis. For example, NCHW16c can describe a 5-D tensor of [batch_size, channel, height, width, channel_block]. Here subordinate axis channel_block=16 is the factor size of the primal axis C (channel). Layout for scalar is defined, while both its name and axes have size 0 |
CLE | Managed reference to LENode |
CLENode | <= b |
CLet | Managed reference to LetNode |
CLetNode | Let binding. Bind var to value then evaluate body |
CLetStmt | Managed reference to LetStmtNode |
CLetStmtNode | Let binding, bind var to value, then run body |
CLoopRV | Managed reference to LoopRVNode |
CLoopRVNode | A random variable that evaluates to a TensorIR for loop |
CLT | Managed reference to LTNode |
CLTNode | < b |
CMatchBufferRegion | Managed reference to MatchBufferRegionNode |
CMatchBufferRegionNode | Match introduces a constraint that the source buffer region can be remapped to the data layout specified by the buffer field. The constraint can be checked in later part of lowering (or optionally during runtime) |
CMax | Managed reference to MaxNode |
CMaxNode | Max(a, b) |
CMemCpyDetails | Helper struct for return value of IdentifyMemCpy |
CMin | Managed reference to MinNode |
CMinNode | Min(a, b) |
CMod | Managed reference to ModNode |
CModNode | % b in the C semnatics |
CMul | Managed reference to MulNode |
CMulNode |
|
CNE | Managed reference to NENode |
CNENode | != b |
CNot | Managed reference to NotNode |
CNotNode | !a |
COr | Managed reference to OrNode |
COrNode | || b |
CPrefetch | Managed reference to PrefetchNode |
CPrefetchNode | A prefetch hint for a buffer |
CPrimFunc | Managed reference to PrimFuncNode |
CPrimFuncNode | Primitive functions that contains TIR statements |
CProducerLoad | Managed reference to ProducerLoadNode |
CProducerLoadNode | Load value from the result produced by the producer |
CProducerRealize | Managed reference to ProducerRealizeNode |
CProducerRealizeNode | Annotate the bounds where the data produced by the producer need to be written and read in body. We will need to allocate space for the corresponding regions |
CProducerStore | Managed reference to ProducerStoreNode |
CProducerStoreNode | Store value into mult-dimensional array that will be read by the consumer of the producer |
CRamp | Managed reference to RampNode |
CRampNode | Construct a vector with lanes elements where its i-th element equals base + i * stride. This is useful to construct a index for a continuous vector load |
CReduce | Managed reference to ReduceNode |
CReduceNode | Reduction operator operator |
CSchedule | Managed reference to ScheduleNode |
CScheduleNode | The user-facing schedule class |
CScheduleState | Managed reference to ScheduleStateNode |
CScheduleStateNode | The state of scheduling, which exposes a Replace method as the primary interface for all the scheduling primitives to manipulate the TensorIR |
CSelect | Managed reference to SelectNode |
CSelectNode | Return true_value if condition is true, otherwise return false_value |
►CSeqStmt | Sequence statement |
CFlattener | Helper class to flatten sequence of arguments into Array |
CSeqStmtNode | The container of seq statement. Represent a sequence of statements |
CShuffle | Managed reference to ShuffleNode |
CShuffleNode | Shuffle instruction. vec = concat(vectors) result = (vec[indices[0]], vec[indices[1]] ...) |
CSizeVar | Named variable represents a tensor index size |
CSizeVarNode | A variable node represent a tensor index size, whose value must be non-negative |
CStmt | Container of all statements |
CStmtExprMutator | Mutator that recursively mutates stmts and exprs on them |
CStmtExprVisitor | Visitor that recursively visit stmts and exprs on them |
CStmtFunctor | Same as ExprFunctor except it is applied on statements |
CStmtFunctor< R(const Stmt &n, Args... args)> | |
CStmtMutator | StmtMutator that mutates the statements |
CStmtNode | Base node of all statements |
CStmtSRef | Managed reference to StmtSRefNode |
CStmtSRefNode | An object that refers to schedulable elements (block/for-loop) in TensorIR, aka "sref" |
CStmtVisitor | StmtVisitor |
CStringImm | Managed reference to StringImmNode |
CStringImmNode | String constants, only used in asserts |
CSub | Managed reference to SubNode |
CSubNode |
|
CTensorIntrin | Managed reference to TensorIntrinNode |
CTensorIntrinNode | Tensor intrinsics for tensorization |
CTrace | Managed reference to TraceNode |
CTraceNode | An execution trace of a scheduling program |
CVar | Named variable in TIR |
CVarNode | A variable node in the IR |
CWhile | Managed reference to WhileNode |
CWhileNode | A While loop |
►Ntopi | |
CEinsumEquation | |
►Ntransform | |
CPass | |
CPassContext | PassContext that is used to configure the pass behavior |
CPassContextNode | PassContextNode contains the information that a pass can rely on, such as analysis results |
CPassInfo | Managed reference class for PassInfoNode |
CPassInfoNode | Meta data that will be used to help optimization and analysis |
CPassNode | PassNode is the base type of differnt types of optimization passes. It is designed as a pure class and implemented by different pass subclasses at different granularity of Relay nodes |
CSequential | |
CSequentialNode | The SequentialNode contains a set of passes that transform Relay programs from one AST to another semantically equivalent one |
CAffineType | Managed reference to AffineTypeNode |
CAffineTypeNode | AffineType representation |
CArrayIndexPath | |
CArrayIndexPathNode | |
CAttrError | Error thrown during attribute checking |
CAttrFieldInfo | AttrFieldInfo |
CAttrFieldInfoNode | Information about attribute fields in string representations |
CAttributeAccessPath | |
CAttributeAccessPathNode | |
CAttrRegistry | |
CAttrRegistryMap | Map<Key, ValueType> used to store meta-data |
CAttrRegistryMapContainerMap | Generic attribute map |
CAttrs | Managed reference to BaseAttrsNode |
CAttrsNode | The base class of the all the Use "curiously recurring template pattern" |
CAttrVisitor | Visitor class to get the attributes of an AST/IR node. The content is going to be called for each field |
CBaseAttrsNode | Base class of all attribute class |
CBaseExpr | Managed reference to BaseExprNode |
CBaseExprNode | Base type of all the expressions |
CBaseFunc | Managed reference to BaseFuncNode |
CBaseFuncNode | Base node of all functions |
CBaseTensorType | Managed reference to BaseTensorTypeNode |
CBaseTensorTypeNode | Base of all Tensor types This container can hold TensorType or GenericTensorType |
CBaseValueEqual | Equality definition of base value class |
CBaseValueHash | Hash definition of base value classes |
CBool | Boolean constant |
CCompilationConfig | Managed reference class to CompilationConfig |
CCompilationConfigNode | Gathers the Targets and distinguished VirtualDevices in canonical form needed to compile a Relay module for execution over possibly heterogeneous devices. Centralizes the validation and canonicalization logic needed to transition from targets supplied by the Python APIs to a single internal representation. Also holds a cache of canonical VirtualDevices so that structural equal virtual devices have pointer equal canonical virtual devices |
CConstantInfo | |
CConstantInfoNode | |
CConstantMemoryPools | |
CConstantMemoryPoolsNode | |
CConstantPoolInfo | |
CConstantPoolInfoNode | |
CConstructor | Managed reference to ConstructorNode |
CConstructorNode | ADT constructor. Constructors compare by pointer equality |
CDiagnostic | |
CDiagnosticBuilder | A wrapper around std::stringstream to build a diagnostic |
CDiagnosticContext | |
CDiagnosticContextNode | |
CDiagnosticNode | A compiler diagnostic message |
CDiagnosticRenderer | |
CDiagnosticRendererNode | Display diagnostics in a given display format |
CDictAttrs | Managed reference to DictAttrsNode |
CDictAttrsNode | Specialized attribute type that is backed by a map. The DictAttrsNode implements the Attrs behavior, its fields are directly accessible via object.field_name like other normal nodes |
CEnvFunc | Managed reference to EnvFuncNode |
CEnvFuncNode | A serializable function backed by TVM's global environment |
CFloatImm | Managed reference class to FloatImmNode |
CFloatImmNode | Constant floating point literals in the program |
CFuncType | Managed reference to FuncTypeNode |
CFuncTypeNode | Function type |
CGenericFunc | Generic function that can be specialized on a per-target basis |
CGenericFuncNode | Represents a generic function that can be specialized on a per-target basis |
CGlobalTypeVar | Managed reference to GlobalTypeVarNode |
CGlobalTypeVarNode | A global type variable that is used for defining new types or type aliases |
CGlobalVar | Managed reference to GlobalVarNode |
CGlobalVarNode | Global variable that lives in the top-level module |
CGlobalVarSupply | Managed reference class to GlobalVarSupplyNode |
CGlobalVarSupplyNode | GlobalVarSupply can be used to generate unique GlobalVars |
CIncompleteType | Managed reference to IncompleteTypeNode |
CIncompleteTypeNode | Intermediate values that is used to indicate incomplete type during type inference |
CInteger | Container of constant int that adds more constructors |
CIntImm | Managed reference class to IntImmNode |
CIntImmNode | Constant integer literals in the program |
CIRModule | Managed reference class to IRModuleNode |
CIRModuleNode | IRModule that holds functions and type definitions |
CMapValuePath | |
CMapValuePathNode | |
CMemoryInfo | Defines memory info |
CMemoryInfoNode | Memory information of special memory region. Use MemoryInfo as its container type |
CMissingArrayElementPath | |
CMissingArrayElementPathNode | |
CMissingMapEntryPath | |
CMissingMapEntryPathNode | |
CNameSupply | Managed reference class to NameSupplyNode |
CNameSupplyNode | NameSupply can be used to generate unique names |
CNDArrayContainerTrait | |
CNodeFunctor | A dynamically dispatched functor on the type of the first argument |
CNodeFunctor< R(const ObjectRef &n, Args...)> | |
CObjectPath | |
CObjectPathNode | Path to an object from some root object |
CObjectPathPair | |
CObjectPathPairNode | Pair of ObjectPath s, one for each object being tested for structural equality |
COp | Managed reference class to OpNode |
COpAttrMap | Map<Op,ValueType> used to store meta-information about Op |
COpNode | Primitive Op(builtin intrinsics) |
COpRegEntry | Helper structure to register operators |
CPointerType | |
CPointerTypeNode | Low-level raw pointer type |
CPoolInfo | Base class for WorkspacePoolInfo and ConstantPoolInfo |
CPoolInfoNode | Describes a pool of memory accessible by one or more targets |
CPoolInfoProperties | |
CPoolInfoPropertiesNode | Describes a pool of memory properties |
CPrimExpr | Reference to PrimExprNode |
CPrimExprNode | Base node of all primitive expressions |
CPrimType | |
CPrimTypeNode | Primitive data types used in the low-level IR |
CPrinterConfig | |
CPrinterConfigNode | |
CRange | Range constainer |
CRangeNode | Range over one dimension |
►CReflectionVTable | Virtual function table to support IR/AST node reflection |
CRegistry | Registry of a reflection table |
CRelayExpr | Managed reference to RelayExprNode |
CRelayExprNode | Base node of all non-primitive expressions |
CRelayRefType | Managed reference to RelayRefTypeNode |
CRelayRefTypeNode | Reference Type High-level Relay IR |
CReprLegacyPrinter | Legacy behavior of ReprPrinter |
CReprPrinter | A printer class to print the AST/IR nodes |
CRootPath | |
CRootPathNode | |
CSEqualHandlerDefault | The default handler for equality testing |
►CSEqualReducer | A Reducer class to reduce the structural equality result of two objects |
CHandler | Internal handler that defines custom behaviors. |
CSHashHandlerDefault | The default handler for hash key computation |
►CSHashReducer | A Reducer class to reduce the structural hash value |
CHandler | Internal handler that defines custom behaviors |
CSource | |
CSourceMap | |
CSourceMapNode | Stores locations in frontend source that generated a node |
CSourceName | The source name of a file span |
CSourceNameNode | The name of a source fragment |
CSourceNode | |
CSpan | |
CSpanNode | Stores locations in frontend source that generated a node |
CStructuralEqual | Content-aware structural equality comparator for objects |
CStructuralHash | Content-aware structural hashing |
CTarget | Managed reference class to TargetNode |
CTargetKind | Managed reference class to TargetKindNode |
CTargetKindAttrMap | Map<TargetKind, ValueType> used to store meta-information about TargetKind |
CTargetKindNode | Target kind, specifies the kind of the target |
CTargetKindRegEntry | Helper structure to register TargetKind |
CTargetNode | Compilation target |
CTargetTag | Managed reference class to TargetTagNode |
CTargetTagNode | A target tag |
CTargetTagRegEntry | |
CTensorAffineType | Managed reference to AffineTypes |
CTensorAffineTypeNode | TensorAffineType representation |
CTensorType | Managed reference to TensorTypeNode |
CTensorTypeNode | This is the most commonly used type in relay. TensorType have a fixed dimension, data type |
CTupleAffineType | Managed reference to TupleAffineTypes |
CTupleAffineTypeNode | TupleAffineType representation |
CTupleType | Managed reference to TupleTypeNode |
CTupleTypeNode | The type of tuple values |
CTVMScriptPrinter | Legacy behavior of ReprPrinter |
CType | Managed reference to TypeNode |
CTypeCall | Managed reference to TypeCallNode |
CTypeCallNode | Type function application |
CTypeConstraint | Managed reference to TypeConstraintNode |
CTypeConstraintNode | Potential Constraints in a function |
CTypeData | Stores all data for an Algebraic Data Type (ADT) |
CTypeDataNode | TypeData container node |
CTypedEnvFunc | Please refer to TypedEnvFunc<R(Args..)> |
CTypedEnvFunc< R(Args...)> | A typed version of EnvFunc. It is backed by a GlobalFuncNode internally |
CTypeFunctor | |
CTypeFunctor< R(const Type &n, Args...)> | |
CTypeMutator | TypeMutator that mutates expressions |
CTypeNode | Type is the base type of all types |
CTypeRelation | Managed reference to TypeRelationNode |
CTypeRelationNode | User defined type relation, it is an input-output relation on types |
CTypeReporter | Container class of TypeReporter |
CTypeReporterNode | Reporter that reports back to the type resolution information |
CTypeVar | Managed reference to TypeVarNode |
CTypeVarNode | Type parameter in functions |
CTypeVisitor | A type visitor that recursively visit types |
CUnknownAttributeAccessPath | |
CUnknownAttributeAccessPathNode | |
CVirtualDevice | Managed reference class to VirtualDeviceNode |
CVirtualDeviceCache | A cache of VirtualDevices . This can be used: |
CVirtualDeviceNode | Describes at compile time the constraints on where data is to be stored at runtime down to the (virtual) device and memory scope level, and how to compile code to compute that data. Used by the PlanDevices pass to collect and solve (virtual) device constraints for the whole Relay program |
CWith | RAII wrapper function to enter and exit a context object similar to python's with syntax |
CWorkspaceMemoryPools | |
CWorkspaceMemoryPoolsNode | |
CWorkspacePoolInfo | |
CWorkspacePoolInfoNode | |
CGlobalVar | |
CMemoryManagerInterface | |
Ctvm_workspace_t | |
CTVMAotExecutor | |
CTVMArgs | |
CTVMByteArray | Byte array type used to pass in byte array When kTVMBytes is used as data type |
CTVMConstantInfo | Describes one constant argument to run_model |
CTVMFuncRegistry | A data structure that facilitates function lookup by C-string name |
CTVMGraphExecutorGraphAttr | |
CTVMMetadata | Top-level metadata structure. Holds all other metadata types |
CTVMModule | Module container of TVM |
CTVMMutableFuncRegistry | A TVMFuncRegistry that supports adding and changing the functions |
CTVMOpParam | Operator attributes about tvm op |
CTVMPackedFunc | |
CTVMParallelGroupEnv | Environment for TVM parallel task |
CTVMTensorInfo | Describes one tensor argument to run_model . NOTE: while TIR allows for other types of arguments, such as scalars, the AOT run_model function does not currently accept these. Therefore it's not possible to express those in this metadata. A future patch may modify this |
CTVMValue | Union type of values being passed through API and function calls |