Ctvm::runtime::memory::Allocator | |
Ctvm::arith::Analyzer | Analyzer that contains bunch of sub-analyzers |
Ctvm::runtime::relax_vm::Instruction::Arg | The auxiliary data structure for instruction argument |
Ctvm::runtime::metadata::ArrayAccessor< C, Ref > | 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 |
Ctvm::runtime::metadata::ArrayAccessor< 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 |
Ctvm::runtime::SimpleObjAllocator::ArrayHandler< ArrayType, ElemType > | |
Ctvm::runtime::metadata::ArrayIterator< C, Ref > | An iterator implementation that lazily instantiates the C++ wrapping Metadata class |
Ctvm::detail::AttrDocEntry | |
Ctvm::detail::AttrDocVisitor | |
Ctvm::detail::AttrExistVisitor | |
Ctvm::detail::AttrInitEntry< T > | |
Ctvm::detail::AttrInitVisitor< FFind > | |
Ctvm::detail::AttrNonDefaultVisitor | |
Ctvm::detail::AttrNopEntry | |
Ctvm::detail::AttrNormalVisitor | |
Ctvm::AttrRegistry< typename, typename > | |
Ctvm::AttrRegistryMap< KeyType, ValueType > | Map<Key, ValueType> used to store meta-data |
►Ctvm::AttrRegistryMap< Op, ValueType > | |
Ctvm::OpAttrMap< ValueType > | Map<Op,ValueType> used to store meta-information about Op |
►Ctvm::AttrRegistryMap< TargetKind, ValueType > | |
Ctvm::TargetKindAttrMap< ValueType > | Map<TargetKind, ValueType> used to store meta-information about TargetKind |
Ctvm::AttrRegistryMapContainerMap< KeyType > | Generic attribute map |
Ctvm::detail::AttrsSEqualVisitor | |
Ctvm::detail::AttrsSHashVisitor | |
Ctvm::detail::AttrTriggerNonDefaultEntry< T > | |
Ctvm::AttrVisitor | Visitor class to get the attributes of an AST/IR node. The content is going to be called for each field |
Ctvm::relax::distributed::Axis | Tensor axis |
Ctvm::relax::distributed::AxisGroupGraph | A graph whose nodes are tensor axes, and the edge means some information can be propagated through the two axes. Although it only does sharding propagation, this data structure can be extended to perform all kinds of propagation that happens on tensor axes |
Ctvm::relax::distributed::AxisGroupHash | |
Ctvm::relax::distributed::AxisHash | |
Ctvm::relax::distributed::AxisShardingSpecEqual | |
Ctvm::relax::distributed::AxisShardingSpecHash | |
►Ctvm::BaseValueEqual | Equality definition of base value class |
Ctvm::StructuralEqual | Content-aware structural equality comparator for objects |
►Ctvm::BaseValueHash | Hash definition of base value classes |
Ctvm::StructuralHash | Content-aware structural hashing |
Ctvm::tir::BlockInfo | 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 |
►Cstd::bool_constant | |
►Ctvm::runtime::is_valid_iterator< T, IterType > | Helper struct for type-checking |
Ctvm::runtime::is_valid_iterator< Optional< T >, IterType > | |
Ctvm::runtime::memory::Buffer | |
Ctvm::tir::BufferAxisHash | |
Ctvm::ir::CalleeCollector | |
Ctvm::runtime::profiling::CallFrame | |
Ctvm::arith::CanonicalSimplifier | Canonical-form based simplifier |
Ctvm::arith::ConstIntBoundAnalyzer | Analyzer to get constant integer bound over expression |
Ctvm::arith::ConstraintContext | Constraint context |
►Ctvm::runtime::NDArray::ContainerBase | The container base structure contains all the fields except for the Object header |
Ctvm::runtime::NDArray::Container | Object container class that backs NDArray |
Ctvm::runtime::DataType | Runtime primitive data type |
Ctvm::runtime::DeviceAPI | TVM Runtime Device API, abstracts the device specific interface for memory management |
Ctvm::relax::DFPatternFunctor< FType > | A dynamical functor that dispatches on in the first DFPattern argument |
►Ctvm::relay::DFPatternFunctor< FType > | A dynamical functor that dispatches on in the first DFPattern argument |
Ctvm::relay::DFPatternVisitor | A simple visitor wrapper around DFPatternFunctor. Recursively visit the content |
Ctvm::relax::DFPatternFunctor< R(const DFPattern &n, Args...)> | |
Ctvm::relay::DFPatternFunctor< R(const DFPattern &n, Args...)> | |
►Ctvm::relax::DFPatternFunctor< void(const DFPattern &)> | |
Ctvm::relax::DFPatternVisitor | A simple visitor wrapper around DFPatternFunctor. Recursively visit the content |
Ctvm::DiagnosticBuilder | A wrapper around std::stringstream to build a diagnostic |
Ctvm::relax::BlockBuilder::DisableOperatorSpecificNormalizationForTVMScript | A marker struct to disable FNormalize |
Ctvm::runtime::DiscoChannel | A bi-directional channel for controler-worker communication. This channel is primarily used to transfer control messages but not data |
Ctvm::runtime::DiscoWorker | A worker in Disco. It takes a channel to communication with the controler. The worker can be run in a separate thread or process as long as the channel supports bi-directional communication in-between |
Ctvm::topi::EinsumEquation | |
►CError | |
Ctvm::AttrError | Error thrown during attribute checking |
Ctvm::relay::CompileError | Custom Error class to be thrown during compilation |
Ctvm::relay::ErrorBuilder | A wrapper around std::stringstream to build error. include/tvm/ir/type.h Can be consumed by CompileError to construct an error |
Ctvm::relay::ErrorReporter | 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 |
Ctvm::relay::ExecutorRegEntry | Helper structure to register Executors |
Ctvm::tir::ExprDeepEqual | Compare two expressions recursively and check if they are equal to each other without var remapping |
Ctvm::relax::ExprFunctor< FType > | 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 |
►Ctvm::relay::ExprFunctor< FType > | 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 |
►Ctvm::relay::ExprMutator | A wrapper around ExprFunctor which functionally updates the AST |
Ctvm::relay::MixedModeMutator | Non-recursive DFS Graph Traversal for Custom Rewriting Passes |
Ctvm::tir::ExprFunctor< FType > | 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 |
►Ctvm::relax::ExprFunctor< Expr(const Expr &)> | |
►Ctvm::relax::ExprMutatorBase | A mutator works in unnormalized form |
Ctvm::relax::ExprMutator | A mutator works in normal form |
►Ctvm::tir::ExprFunctor< PrimExpr(const PrimExpr &)> | |
►Ctvm::tir::ExprMutator | ExprMutator that mutates expressions |
►Ctvm::tir::StmtExprMutator | Mutator that recursively mutates stmts and exprs on them |
►Ctvm::tir::DataTypeLegalizer | Legalize the data types of expressions to make sure they are consistent with other parts of the program |
►Ctvm::tir::IndexDataTypeRewriter | Data type rewriter for buffer indices |
Ctvm::tir::IndexDataTypeNormalizer | Normalize the data types of buffer shapes and indices to the same data type |
Ctvm::relax::ExprFunctor< R(const Expr &n, Args...)> | |
Ctvm::relay::ExprFunctor< R(const Expr &n, Args...)> | |
Ctvm::tir::ExprFunctor< R(const PrimExpr &n, Args...)> | |
►Ctvm::relax::ExprFunctor< void(const Expr &)> | |
Ctvm::relax::ExprVisitor | A simple visitor wrapper around ExprFunctor. Recursively visit the content |
►Ctvm::relay::ExprFunctor< void(const Expr &n)> | |
►Ctvm::relay::ExprVisitor | A simple visitor wrapper around ExprFunctor. Recursively visit the content |
Ctvm::relay::MixedModeVisitor | A wrapper around ExprVisitor which traverses the Dataflow Normal AST |
►Ctvm::tir::ExprFunctor< void(const PrimExpr &)> | |
►Ctvm::tir::ExprVisitor | ExprVisitor |
►Ctvm::tir::StmtExprVisitor | Visitor that recursively visit stmts and exprs on them |
Ctvm::tir::BufferAxisGraphExtractor | Construct an axis group graph from a PrimFunc. Two buffer axis are connected if they are accessed by the same index |
Ctvm::relay::ExprRewriter | A non-iterating Expression Rewriter |
Ctvm::runtime::PackedFuncObj::Extractor< TPackedFuncSubObj > | Internal struct for extracting the callable method from callable type |
►Cstd::false_type | |
Ctvm::detail::is_specialized< Type, Container > | |
Ctvm::relay::FeatureSet | A finite set of Feature |
Ctvm::runtime::relax_vm::NDArrayCacheMetadata::FileRecord | Each shard of NDArray cache, which by default, is named as "params_shard_x.bin" |
Ctvm::tir::SeqStmt::Flattener | Helper class to flatten sequence of arguments into Array |
CGlobalVar | |
Ctvm::transform::GlobalVarReplacer | |
Ctvm::tir::usmp::algo::GreedyBase | This is the base class for Greedy Algorithms where the sorting is specialized in the extended classes based on the greedy criteria |
Ctvm::runtime::SimpleObjAllocator::Handler< T > | |
►Ctvm::SEqualReducer::Handler | Internal handler that defines custom behaviors. |
Ctvm::SEqualHandlerDefault | The default handler for equality testing |
►Ctvm::SHashReducer::Handler | Internal handler that defines custom behaviors |
Ctvm::SHashHandlerDefault | The default handler for hash key computation |
Cdmlc::serializer::Handler< DLDataType > | |
Cdmlc::serializer::Handler< DLDevice > | |
►Ctvm::detail::ImplSEqualReduce< T, bool > | |
Ctvm::detail::ReflectionTrait< T > | |
Ctvm::detail::ImplSEqualReduce< T, true > | |
►Ctvm::detail::ImplSHashReduce< T, bool > | |
Ctvm::detail::ReflectionTrait< T > | |
Ctvm::detail::ImplSHashReduce< T, true > | |
►Ctvm::detail::ImplVisitAttrs< T, bool > | |
Ctvm::detail::ReflectionTrait< T > | |
Ctvm::detail::ImplVisitAttrs< T, true > | |
Ctvm::runtime::InplaceArrayBase< ArrayType, ElemType > | Base template for classes with array like memory layout |
►Ctvm::runtime::InplaceArrayBase< ADTObj, ObjectRef > | |
Ctvm::runtime::ADTObj | An object representing a structure or enumeration |
►Ctvm::runtime::InplaceArrayBase< ArrayNode, ObjectRef > | |
Ctvm::runtime::ArrayNode | Array node content in array |
►Ctvm::runtime::InplaceArrayBase< SmallMapNode, MapNode::KVType > | |
Ctvm::runtime::SmallMapNode | A specialization of small-sized hash map |
Ctvm::runtime::relax_vm::Instruction | A single virtual machine instruction |
Ctvm::runtime::vm::Instruction | A single virtual machine instruction |
Ctvm::tir::InstructionKindRegEntry | An entry in the registry of InstructionKind |
Ctvm::arith::IntSetAnalyzer | Integer set analyzer |
Ctvm::script::printer::IRDocsifierFunctor< R, Args > | Dynamic dispatch functor based on ObjectPath |
Ctvm::runtime::IterAdapter< Converter, TIter > | Iterator adapter that adapts TIter to return another type |
Ctvm::runtime::Map< K, V, typename, typename >::iterator | Iterator of the hash map |
Ctvm::runtime::MapNode::iterator | |
Ctvm::support::Span< T, W >::iterator_base< W1 > | |
Ctvm::auto_scheduler::AttachMapNode::IterKeyHash | |
Ctvm::tir::LayoutAxis | |
Ctvm::support::LinearCongruentialEngine | 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 |
Ctvm::tir::MemCpyDetails | Helper struct for return value of IdentifyMemCpy |
Ctvm::runtime::memory::MemoryManager | |
Ctvm::arith::ModularSetAnalyzer | Analyzer to get modular information over expression |
Ctvm::script::ir_builder::details::Namer | |
Ctvm::runtime::relax_vm::NDArrayCacheMetadata | Metadata for NDArray cache, which by default, is named as "ndarray-cache.json" |
Ctvm::NDArrayContainerTrait | |
Ctvm::NodeFunctor< FType > | A dynamically dispatched functor on the type of the first argument |
Ctvm::NodeFunctor< R(const ObjectRef &n, Args...)> | |
Ctvm::runtime::NullOptType | Helper to represent nullptr for optional |
Ctvm::runtime::NVTXScopedRange | A class to create a NVTX range. No-op if TVM is not built against NVTX |
Ctvm::runtime::ObjAllocatorBase< Derived > | Base class of object allocators that implements make. Use curiously recurring template pattern |
►Ctvm::runtime::ObjAllocatorBase< SimpleObjAllocator > | |
Ctvm::runtime::SimpleObjAllocator | |
►Ctvm::runtime::Object | Base class of all object containers |
►Ctvm::AffineTypeNode | AffineType representation |
Ctvm::TensorAffineTypeNode | TensorAffineType representation |
Ctvm::TupleAffineTypeNode | TupleAffineType representation |
Ctvm::AttrFieldInfoNode | Information about attribute fields in string representations |
►Ctvm::BaseAttrsNode | Base class of all attribute class |
►Ctvm::AttrsNode< ReduceAttrs > | |
Ctvm::relay::ReduceAttrs | Attributes for Reduce operators |
►Ctvm::AttrsNode< GridSampleAttrs > | |
Ctvm::relay::GridSampleAttrs | Attributes used in image grid_sample operator |
►Ctvm::AttrsNode< DilateAttrs > | |
Ctvm::relay::DilateAttrs | Attributes used in dilate operator |
►Ctvm::AttrsNode< DeformableConv2DAttrs > | |
Ctvm::relay::DeformableConv2DAttrs | Attributes for DeformableConv2D operator |
►Ctvm::AttrsNode< Conv2DWinogradNNPACKWeightTransformAttrs > | |
Ctvm::relay::Conv2DWinogradNNPACKWeightTransformAttrs | Attributes used in winograd weight transformation operators |
►Ctvm::AttrsNode< BinaryDenseAttrs > | |
Ctvm::relay::BinaryDenseAttrs | |
►Ctvm::AttrsNode< AllClassNonMaximumSuppressionAttrs > | |
Ctvm::relay::AllClassNonMaximumSuppressionAttrs | Attributes used in all_class_non_maximum_suppression operator |
►Ctvm::AttrsNode< ScatterCollectiveAttrs > | |
Ctvm::relax::ScatterCollectiveAttrs | Attributes used in scatter operators |
►Ctvm::AttrsNode< HintOnDeviceAttrs > | |
Ctvm::relax::HintOnDeviceAttrs | Attributes used in hint_on_device |
►Ctvm::AttrsNode< BroadcastAttrs > | |
Ctvm::relay::qnn::BroadcastAttrs | Attribute for broadcast operator |
►Ctvm::AttrsNode< ThreefryGenerateAttrs > | |
Ctvm::relay::ThreefryGenerateAttrs | |
►Ctvm::AttrsNode< SubPixelAttrs > | |
Ctvm::relay::SubPixelAttrs | Attributes used in subpixel operators |
►Ctvm::AttrsNode< SparseDenseAttrs > | |
Ctvm::relay::SparseDenseAttrs | Attributes for sparse_dense operator |
►Ctvm::AttrsNode< ShapeOfAttrs > | |
Ctvm::relay::ShapeOfAttrs | Attributes for ShapeOf operator |
►Ctvm::AttrsNode< MeshgridAttrs > | |
Ctvm::relay::MeshgridAttrs | Attributes used in meshgrid operators |
►Ctvm::AttrsNode< BatchToSpaceNDAttrs > | |
Ctvm::relay::BatchToSpaceNDAttrs | Attributes used in BatchToSpaceND operator |
►Ctvm::AttrsNode< AllocTensorAttrs > | |
Ctvm::relay::AllocTensorAttrs | Options for allocating tensors |
►Ctvm::AttrsNode< TakeAttrs > | |
Ctvm::relax::TakeAttrs | Attributes used in take operator |
Ctvm::relay::TakeAttrs | |
►Ctvm::AttrsNode< ScatterElementsAttrs > | |
Ctvm::relax::ScatterElementsAttrs | Attributes used in scatter_elements operators |
Ctvm::relay::ScatterElementsAttrs | |
►Ctvm::AttrsNode< GatherElementsAttrs > | |
Ctvm::relax::GatherElementsAttrs | Attributes used in gather_elements operators |
►Ctvm::AttrsNode< EinsumAttrs > | |
Ctvm::relax::EinsumAttrs | Attributes used in einsum operator |
Ctvm::relay::EinsumAttrs | Attributes used in einsum operator |
►Ctvm::AttrsNode< DistributionAttrs > | |
Ctvm::relax::DistributionAttrs | Attributes for redistribute and annotate_sharding operator |
►Ctvm::AttrsNode< CallInplacePackedAttrs > | |
Ctvm::relax::CallInplacePackedAttrs | Attributes used in call_inplace_packed |
►Ctvm::AttrsNode< AttentionAttrs > | |
Ctvm::relax::AttentionAttrs | Attributes used in Attention operator |
►Ctvm::AttrsNode< AllReduceAttrs > | |
Ctvm::relax::AllReduceAttrs | Attributes used in allreduce operators |
►Ctvm::AttrsNode< AdaptivePool2DAttrs > | |
Ctvm::relax::AdaptivePool2DAttrs | Attributes for 2d adaptive pool operator |
Ctvm::relay::AdaptivePool2DAttrs | Attributes for 2d adaptive pool operator |
►Ctvm::AttrsNode< NdarraySizeAttrs > | |
Ctvm::relay::NdarraySizeAttrs | Attributes for ndarray_size operator |
►Ctvm::AttrsNode< GatherAttrs > | |
Ctvm::relay::GatherAttrs | |
►Ctvm::AttrsNode< ScatterNDAttrs > | |
Ctvm::relax::ScatterNDAttrs | Attributes used in scatter_nd operators |
Ctvm::relay::ScatterNDAttrs | |
►Ctvm::AttrsNode< InitAttrs > | |
Ctvm::relax::InitAttrs | Attributes used in full/full_like, ones/ones_like, and zeros/zeros_like operators |
►Ctvm::AttrsNode< FlipAttrs > | |
Ctvm::relax::FlipAttrs | Attributes used in flip operators |
►Ctvm::AttrsNode< AllGatherAttrs > | |
Ctvm::relax::AllGatherAttrs | Attributes used in allgather operators |
►Ctvm::AttrsNode< AdaptivePool3DAttrs > | |
Ctvm::relax::AdaptivePool3DAttrs | Attributes for 3d adaptive pool operator |
Ctvm::relay::AdaptivePool3DAttrs | Attributes for 3d adaptive pool operator |
►Ctvm::AttrsNode< RequantizeAttrs > | |
Ctvm::relay::qnn::RequantizeAttrs | Attribute for requantize operator |
►Ctvm::AttrsNode< Resize1DAttrs > | |
Ctvm::relay::Resize1DAttrs | Attributes used in image resize1d operator |
►Ctvm::AttrsNode< OnDeviceAttrs > | |
Ctvm::relay::OnDeviceAttrs | Attributes for the "on_device" annotation (ie operator) |
►Ctvm::AttrsNode< NonMaximumSuppressionAttrs > | |
Ctvm::relay::NonMaximumSuppressionAttrs | Attributes used in non_maximum_suppression operator |
►Ctvm::AttrsNode< MultiBoxPriorAttrs > | |
Ctvm::relay::MultiBoxPriorAttrs | Attributes used in multibox_prior operators |
►Ctvm::AttrsNode< DFTAttrs > | |
Ctvm::relay::DFTAttrs | Attributes used in DFT operator |
►Ctvm::AttrsNode< CropAndResizeAttrs > | |
Ctvm::relay::CropAndResizeAttrs | Attributes used in image crop_and_resize operator |
►Ctvm::AttrsNode< BitPackAttrs > | |
Ctvm::relay::BitPackAttrs | Attributes used in bitpack operators |
►Ctvm::AttrsNode< ToVDeviceAttrs > | |
Ctvm::relax::ToVDeviceAttrs | Attributes used in to_vdevice |
►Ctvm::AttrsNode< ExpandDimsAttrs > | |
Ctvm::relax::ExpandDimsAttrs | Attributes used in expand_dims operators |
Ctvm::relay::ExpandDimsAttrs | Attributes used in expand_dims operators |
►Ctvm::AttrsNode< ConcatAttrs > | |
Ctvm::relax::ConcatAttrs | Attributes used in concat operators |
►Ctvm::AttrsNode< StftAttrs > | |
Ctvm::relay::StftAttrs | Attributes used in stft operator |
►Ctvm::AttrsNode< SparseConv2DAttrs > | |
Ctvm::relay::SparseConv2DAttrs | Attributes for sparse_dense operator |
►Ctvm::AttrsNode< ReshapeLikeAttrs > | |
Ctvm::relay::ReshapeLikeAttrs | Attributes used in MXNet-style reshape_like operators |
►Ctvm::AttrsNode< ROIAlignAttrs > | |
Ctvm::relay::ROIAlignAttrs | Attributes used in roi_align operators |
►Ctvm::AttrsNode< TriluAttrs > | |
Ctvm::relax::TriluAttrs | Attributes used in tril and triu operator |
Ctvm::relay::TriluAttrs | |
►Ctvm::AttrsNode< SoftmaxAttrs > | |
Ctvm::relax::SoftmaxAttrs | Attributes used in softmax operators |
Ctvm::relay::SoftmaxAttrs | Attributes used in softmax operators |
►Ctvm::AttrsNode< AdaptivePool1DAttrs > | |
Ctvm::relax::AdaptivePool1DAttrs | Attributes for 1d adaptive pool operator |
Ctvm::relay::AdaptivePool1DAttrs | Attributes for 1d adaptive pool operator |
►Ctvm::AttrsNode< SpaceToBatchNDAttrs > | |
Ctvm::relay::SpaceToBatchNDAttrs | Attributes used in SpaceToBatchND operator |
►Ctvm::AttrsNode< SlidingWindowAttrs > | |
Ctvm::relay::SlidingWindowAttrs | Attributes used for the sliding_window operator |
►Ctvm::AttrsNode< Resize3DAttrs > | |
Ctvm::relay::Resize3DAttrs | Attributes used in image resize3d operator |
►Ctvm::AttrsNode< FixedPointMultiplyPerAxisAttrs > | |
Ctvm::relay::FixedPointMultiplyPerAxisAttrs | Attributes for per channel/per axes FixedPointMultiply operator |
►Ctvm::AttrsNode< AvgPool1DAttrs > | |
Ctvm::relay::AvgPool1DAttrs | Attributes for 1D avg pool operator |
►Ctvm::AttrsNode< MatmulAttrs > | |
Ctvm::relax::MatmulAttrs | Attributes for matmul operator |
Ctvm::relay::MatmulAttrs | Attributes for matmul operator |
►Ctvm::AttrsNode< ArgmaxArgminAttrs > | |
Ctvm::relax::ArgmaxArgminAttrs | Attributes for search operators |
►Ctvm::AttrsNode< DequantizeAttrs > | |
Ctvm::relay::qnn::DequantizeAttrs | Attribute for dequantize operator |
►Ctvm::AttrsNode< YoloReorgAttrs > | |
Ctvm::relay::YoloReorgAttrs | Attributes used in yolo reorg operators |
►Ctvm::AttrsNode< UniformAttrs > | |
Ctvm::relay::UniformAttrs | |
►Ctvm::AttrsNode< ReshapeTensorAttrs > | |
Ctvm::relay::ReshapeTensorAttrs | Attributes for VM reshape_tensor operator |
►Ctvm::AttrsNode< MultinomialAttrs > | |
Ctvm::relay::MultinomialAttrs | |
►Ctvm::AttrsNode< MetaScheduleLayoutTransformAttrs > | |
Ctvm::relay::MetaScheduleLayoutTransformAttrs | Attributes for MetaScheduleLayoutTransform operator |
►Ctvm::AttrsNode< CompilerAttrs > | |
Ctvm::relay::CompilerAttrs | Options for the operators used to annotate a compiler |
►Ctvm::AttrsNode< Resize2DAttrs > | |
Ctvm::relax::Resize2DAttrs | Attributes used in image resize2d operator |
Ctvm::relay::Resize2DAttrs | Attributes used in image resize2d operator |
►Ctvm::AttrsNode< Pool3DAttrs > | |
Ctvm::relax::Pool3DAttrs | Attributes used in max_pool3d and avg_pool3d operator |
►Ctvm::AttrsNode< LayerNormAttrs > | |
Ctvm::relax::LayerNormAttrs | Attributes used in layer_norm operator |
Ctvm::relay::LayerNormAttrs | Attributes used in layer_norm operator |
►Ctvm::AttrsNode< RegularNonMaximumSuppressionAttrs > | |
Ctvm::relay::RegularNonMaximumSuppressionAttrs | Attributes used in regular_non_maximum_suppression operator |
►Ctvm::AttrsNode< ProposalAttrs > | |
Ctvm::relay::ProposalAttrs | Attributes used in proposal operators |
►Ctvm::AttrsNode< Conv3DWinogradAttrs > | |
Ctvm::relay::Conv3DWinogradAttrs | Attributes used in 3d winograd convolution operators |
►Ctvm::AttrsNode< BiasAddAttrs > | |
Ctvm::relay::BiasAddAttrs | Add a 1D Tensor to an axis of a data |
►Ctvm::AttrsNode< AvgPool3DAttrs > | |
Ctvm::relay::AvgPool3DAttrs | Attributes for 3D avg pool operator |
►Ctvm::AttrsNode< SqueezeAttrs > | |
Ctvm::relax::SqueezeAttrs | Attributes used in squeeze operators |
Ctvm::relay::SqueezeAttrs | Attributes used in squeeze operators |
►Ctvm::AttrsNode< Pool2DAttrs > | |
Ctvm::relax::Pool2DAttrs | Attributes used in max_pool2d and avg_pool2d operator |
►Ctvm::AttrsNode< OneHotAttrs > | |
Ctvm::relax::OneHotAttrs | Attributes used in one_hot operator |
Ctvm::relay::OneHotAttrs | Attributes used in one-hot operator |
►Ctvm::AttrsNode< NLLLossAttrs > | |
Ctvm::relax::NLLLossAttrs | Attributes used in nll_loss operator |
Ctvm::relay::NLLLossAttrs | Attributes used in NLLLoss operator |
►Ctvm::AttrsNode< BatchNormAttrs > | |
Ctvm::relax::BatchNormAttrs | Attributes used in batch_norm operator |
Ctvm::relay::BatchNormAttrs | Attributes used in batch_norm operator |
►Ctvm::AttrsNode< SparseToDenseAttrs > | |
Ctvm::relay::SparseToDenseAttrs | Attributes used in sparse_to_dense operator |
►Ctvm::AttrsNode< SearchSortedAttrs > | |
Ctvm::relay::SearchSortedAttrs | |
►Ctvm::AttrsNode< ReverseAttrs > | |
Ctvm::relay::ReverseAttrs | Attributes used in reverse operators |
►Ctvm::AttrsNode< InitOpAttrs > | |
Ctvm::relay::InitOpAttrs | Attributes that specify a tensor |
►Ctvm::AttrsNode< Conv2DWinogradAttrs > | |
Ctvm::relay::Conv2DWinogradAttrs | Attributes used in convolution operators with winograd algorithm |
►Ctvm::AttrsNode< ConcatenateAttrs > | |
Ctvm::relay::ConcatenateAttrs | Attributes used in concatenate operators |
►Ctvm::AttrsNode< AvgPool2DAttrs > | |
Ctvm::relay::AvgPool2DAttrs | Attributes for avg pool operator |
►Ctvm::AttrsNode< RepeatAttrs > | |
Ctvm::relax::RepeatAttrs | Attributes used in repeat operators |
Ctvm::relay::RepeatAttrs | Attributes used in repeat operators |
►Ctvm::AttrsNode< Pool1DAttrs > | |
Ctvm::relax::Pool1DAttrs | Attributes used in max_pool1d and avg_pool1d operator |
►Ctvm::AttrsNode< TransposeAttrs > | |
Ctvm::relay::TransposeAttrs | Attributes used in transpose operators |
►Ctvm::AttrsNode< StackAttrs > | |
Ctvm::relay::StackAttrs | Attributes used in stack operators |
►Ctvm::AttrsNode< ReverseSequenceAttrs > | |
Ctvm::relay::ReverseSequenceAttrs | Attributes used in reverse_sequence operators |
►Ctvm::AttrsNode< CallLoweredAttrs > | |
Ctvm::relay::CallLoweredAttrs | Metadata for calls to TIR functions, useful for program analysis crossing Relay and TIR |
►Ctvm::AttrsNode< BatchMatmulAttrs > | |
Ctvm::relay::BatchMatmulAttrs | Attributes for batch matmul operator |
►Ctvm::AttrsNode< ArangeAttrs > | |
Ctvm::relay::ArangeAttrs | Attributes used in arange operators |
►Ctvm::AttrsNode< ScanopAttrs > | |
Ctvm::relax::ScanopAttrs | Attributes used in scan operators like cumsum, cumprod |
Ctvm::relay::ScanopAttrs | Attributes used in cumsum and cumprod operator |
►Ctvm::AttrsNode< LeakyReluAttrs > | |
Ctvm::relax::LeakyReluAttrs | Attributes used in softmax operators |
Ctvm::relay::LeakyReluAttrs | Attributes for leaky relu operator |
►Ctvm::AttrsNode< CallTIRWithGradAttrs > | |
Ctvm::relax::CallTIRWithGradAttrs | Attributes used in call_tir_with_grad |
►Ctvm::AttrsNode< CallTIRInplaceAttrs > | |
Ctvm::relax::CallTIRInplaceAttrs | Attributes used in call_tir_inplace |
►Ctvm::AttrsNode< AstypeAttrs > | |
Ctvm::relax::AstypeAttrs | Attributes used in astype operator |
►Ctvm::AttrsNode< LRNAttrs > | |
Ctvm::relay::LRNAttrs | Attributes for LRN operator |
►Ctvm::AttrsNode< InstanceNormAttrs > | |
Ctvm::relay::InstanceNormAttrs | Attributes used in instance_norm operator |
►Ctvm::AttrsNode< FIFOBufferAttrs > | |
Ctvm::relay::FIFOBufferAttrs | Attributes for FIFO buffer operator |
►Ctvm::AttrsNode< PadAttrs > | |
Ctvm::relax::PadAttrs | Attributes used for the padding operator |
Ctvm::relay::PadAttrs | Attributes used for the padding operator |
►Ctvm::AttrsNode< DropoutAttrs > | |
Ctvm::relax::DropoutAttrs | Attributes used in dropout operator |
Ctvm::relay::DropoutAttrs | Attributes used in dropout operator |
►Ctvm::AttrsNode< SliceLikeAttrs > | |
Ctvm::relay::SliceLikeAttrs | |
►Ctvm::AttrsNode< ShapeFuncAttrs > | |
Ctvm::relay::ShapeFuncAttrs | Options for the shape function operator |
►Ctvm::AttrsNode< ReshapeAttrs > | |
Ctvm::relay::ReshapeAttrs | Attributes used in reshape operators |
►Ctvm::AttrsNode< ROIPoolAttrs > | |
Ctvm::relay::ROIPoolAttrs | Attributes used in roi_pool operators |
►Ctvm::AttrsNode< DynExpandDimsAttrs > | |
Ctvm::relay::DynExpandDimsAttrs | Attributes used in dynamic expand_dims operators |
►Ctvm::AttrsNode< CastHintAttrs > | |
Ctvm::relay::CastHintAttrs | Annotate an expression to be cast into specific data type |
►Ctvm::AttrsNode< BinaryConv2DAttrs > | |
Ctvm::relay::BinaryConv2DAttrs | Attribues used in bitserial convolution operators |
►Ctvm::AttrsNode< TopKAttrs > | |
Ctvm::relax::TopKAttrs | Attributes used in topk operator |
Ctvm::relay::TopKAttrs | |
►Ctvm::AttrsNode< MultinomialFromUniformAttrs > | |
Ctvm::relax::MultinomialFromUniformAttrs | Attributes used in multinomial_from_uniform operator |
►Ctvm::AttrsNode< Conv3DAttrs > | |
Ctvm::relax::Conv3DAttrs | Attributes used in Conv3d operator |
Ctvm::relay::Conv3DAttrs | Attributes used in convolution operators |
►Ctvm::AttrsNode< VirtualDeviceNode > | |
Ctvm::VirtualDeviceNode | 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 |
►Ctvm::AttrsNode< SparseTransposeAttrs > | |
Ctvm::relay::SparseTransposeAttrs | Attributes for sparse_transpose operator |
►Ctvm::AttrsNode< PReluAttrs > | |
Ctvm::relay::PReluAttrs | Attributes for prelu operator |
►Ctvm::AttrsNode< MaxPool1DAttrs > | |
Ctvm::relay::MaxPool1DAttrs | Attributes for 1D max pool operator |
►Ctvm::AttrsNode< L2NormalizeAttrs > | |
Ctvm::relay::L2NormalizeAttrs | Attributes for L2Normalize operator |
►Ctvm::AttrsNode< ConvGemmWeightTransformAttrs > | |
Ctvm::relay::ConvGemmWeightTransformAttrs | Attributes used in gemm weight transformation operators |
►Ctvm::AttrsNode< ArgReduceAttrs > | |
Ctvm::relay::ArgReduceAttrs | Attributes for Reduce operators which reduce by finding a single element. E.g. argmin |
►Ctvm::AttrsNode< WrapParamAttrs > | |
Ctvm::relax::WrapParamAttrs | Attributes used in wrap_param operator |
►Ctvm::AttrsNode< TileAttrs > | |
Ctvm::relax::TileAttrs | Attributes used in tile operators |
Ctvm::relay::TileAttrs | Attributes used in tile operators |
►Ctvm::AttrsNode< StridedSliceAttrs > | |
Ctvm::relax::StridedSliceAttrs | Attributes used in strided_slice operator |
Ctvm::relay::StridedSliceAttrs | Attributes for StridedSlice operator |
►Ctvm::AttrsNode< SplitAttrs > | |
Ctvm::relax::SplitAttrs | Attributes used in split operator |
Ctvm::relay::SplitAttrs | |
►Ctvm::AttrsNode< RMSNormAttrs > | |
Ctvm::relax::RMSNormAttrs | Attributes used in rms_norm operator |
►Ctvm::AttrsNode< QuantizeAttrs > | |
Ctvm::relax::QuantizeAttrs | Attributes for relax.quantize/relax.dequantize operator |
Ctvm::relay::qnn::QuantizeAttrs | Attribute for quantize operator |
►Ctvm::AttrsNode< Conv2DAttrs > | |
Ctvm::relax::Conv2DAttrs | Attributes used in Conv2d operator |
Ctvm::relay::Conv2DAttrs | Attributes used in convolution operators |
►Ctvm::AttrsNode< SimulatedQuantizeAttrs > | |
Ctvm::relay::qnn::SimulatedQuantizeAttrs | |
►Ctvm::AttrsNode< UpSampling3DAttrs > | |
Ctvm::relay::UpSampling3DAttrs | Attributes for upsampling3d operator |
►Ctvm::AttrsNode< MirrorPadAttrs > | |
Ctvm::relay::MirrorPadAttrs | Attributes used for the MirrorPadding operator |
►Ctvm::AttrsNode< GlobalPool2DAttrs > | |
Ctvm::relay::GlobalPool2DAttrs | Attributes for global pool operator |
►Ctvm::AttrsNode< FixedPointMultiplyAttrs > | |
Ctvm::relay::FixedPointMultiplyAttrs | Attributes for FixedPointMultiply operator |
►Ctvm::AttrsNode< DeviceCopyAttrs > | |
Ctvm::relay::DeviceCopyAttrs | Options for the device copy operators |
►Ctvm::AttrsNode< CastAttrs > | |
Ctvm::relay::CastAttrs | Data type cast |
►Ctvm::AttrsNode< AllocStorageAttrs > | |
Ctvm::relay::AllocStorageAttrs | Options for allocating storage |
►Ctvm::AttrsNode< SortAttrs > | |
Ctvm::relax::SortAttrs | Attributes used in sort operator |
►Ctvm::AttrsNode< Conv1DAttrs > | |
Ctvm::relax::Conv1DAttrs | Attributes used in Conv1d operator |
Ctvm::relay::Conv1DAttrs | Attributes used in 1D convolution operators |
►Ctvm::AttrsNode< ArgsortAttrs > | |
Ctvm::relax::ArgsortAttrs | Attributes used in argsort operator |
Ctvm::relay::ArgsortAttrs | Attributes used in argsort operators |
►Ctvm::AttrsNode< SequenceMaskAttrs > | |
Ctvm::relay::SequenceMaskAttrs | |
►Ctvm::AttrsNode< MultiBoxTransformLocAttrs > | |
Ctvm::relay::MultiBoxTransformLocAttrs | |
►Ctvm::AttrsNode< MaxPool3DAttrs > | |
Ctvm::relay::MaxPool3DAttrs | Attributes for 3D max pool operator |
►Ctvm::AttrsNode< MatrixSetDiagAttrs > | |
Ctvm::relay::MatrixSetDiagAttrs | Attributes used in matrix_set_diag operator |
►Ctvm::AttrsNode< DensePackAttrs > | |
Ctvm::relay::DensePackAttrs | Attributes for dense_pack operator |
►Ctvm::AttrsNode< DenseAttrs > | |
Ctvm::relay::DenseAttrs | Attributes for dense operator |
►Ctvm::AttrsNode< ClipAttrs > | |
Ctvm::relay::ClipAttrs | Attributes for Clip operator |
►Ctvm::AttrsNode< AutoSchedulerLayoutTransformAttrs > | |
Ctvm::relay::AutoSchedulerLayoutTransformAttrs | Attributes for AutoSchedulerLayoutTransform operator |
►Ctvm::AttrsNode< StatisticalAttrs > | |
Ctvm::relax::StatisticalAttrs | Attributes for statistical operators |
►Ctvm::AttrsNode< LayoutTransformAttrs > | |
Ctvm::relax::LayoutTransformAttrs | Attributes used in layout_transform operator |
Ctvm::relay::LayoutTransformAttrs | Attributes for LayoutTransform operator |
►Ctvm::AttrsNode< Conv1DTransposeAttrs > | |
Ctvm::relax::Conv1DTransposeAttrs | Attributes used in Conv1DTranspose operator |
Ctvm::relay::Conv1DTransposeAttrs | Attributes used in 1D transposed convolution operator |
►Ctvm::AttrsNode< VarianceAttrs > | |
Ctvm::relay::VarianceAttrs | |
►Ctvm::AttrsNode< UniqueAttrs > | |
Ctvm::relay::UniqueAttrs | Attributes used in unique operator |
►Ctvm::AttrsNode< MaxPool2DAttrs > | |
Ctvm::relay::MaxPool2DAttrs | Attributes for max pool operator |
►Ctvm::AttrsNode< GetValidCountsAttrs > | |
Ctvm::relay::GetValidCountsAttrs | Attributes used in get_valid_counts operator |
►Ctvm::AttrsNode< DebugAttrs > | |
Ctvm::relay::DebugAttrs | Options for the debug operators |
►Ctvm::AttrsNode< CorrelationAttrs > | |
Ctvm::relay::CorrelationAttrs | Attributes used in correlation operators |
►Ctvm::AttrsNode< ConvWinogradWeightTransformAttrs > | |
Ctvm::relay::ConvWinogradWeightTransformAttrs | Attributes used in winograd weight transformation operators |
►Ctvm::AttrsNode< Conv2DTransposeAttrs > | |
Ctvm::relax::Conv2DTransposeAttrs | Attributes used in Conv2d operator |
Ctvm::relay::Conv2DTransposeAttrs | Attributes used in transposed convolution operator |
►Ctvm::AttrsNode< UpSamplingAttrs > | |
Ctvm::relay::UpSamplingAttrs | Attributes for upsampling operator |
►Ctvm::AttrsNode< NormalAttrs > | |
Ctvm::relay::NormalAttrs | |
►Ctvm::AttrsNode< Dilation2DAttrs > | |
Ctvm::relay::Dilation2DAttrs | Attributes used in dilation operators |
►Ctvm::AttrsNode< Conv3DTransposeAttrs > | |
Ctvm::relay::Conv3DTransposeAttrs | Attributes used in transposed convolution operator |
►Ctvm::AttrsNode< AffineGridAttrs > | |
Ctvm::relay::AffineGridAttrs | Attributes used in image affine_grid operator |
►Ctvm::AttrsNode< PermuteDimsAttrs > | |
Ctvm::relax::PermuteDimsAttrs | Attributes used in permute_dims operator |
►Ctvm::AttrsNode< GroupNormAttrs > | |
Ctvm::relax::GroupNormAttrs | Attributes used in group_norm operator |
Ctvm::relay::GroupNormAttrs | Attributes used in group_norm operator |
►Ctvm::AttrsNode< GatherNDAttrs > | |
Ctvm::relax::GatherNDAttrs | Attributes used in gather_nd operators |
Ctvm::relay::GatherNDAttrs | |
Ctvm::AttrsNode< DerivedType > | The base class of the all the Use "curiously recurring template pattern" |
Ctvm::DictAttrsNode | 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 |
►Ctvm::BaseExprNode | Base type of all the expressions |
►Ctvm::PrimExprNode | Base node of all primitive expressions |
►Ctvm::tir::BinaryOpNode< AddNode > | |
Ctvm::tir::AddNode |
|
►Ctvm::tir::BinaryOpNode< MinNode > | |
Ctvm::tir::MinNode | Min(a, b) |
►Ctvm::tir::BinaryOpNode< DivNode > | |
Ctvm::tir::DivNode | / b in the C semnatics |
►Ctvm::tir::BinaryOpNode< FloorModNode > | |
Ctvm::tir::FloorModNode | The remainder of the floordiv |
►Ctvm::tir::BinaryOpNode< MulNode > | |
Ctvm::tir::MulNode |
|
►Ctvm::tir::BinaryOpNode< ModNode > | |
Ctvm::tir::ModNode | % b in the C semnatics |
►Ctvm::tir::BinaryOpNode< MaxNode > | |
Ctvm::tir::MaxNode | Max(a, b) |
►Ctvm::tir::BinaryOpNode< FloorDivNode > | |
Ctvm::tir::FloorDivNode | Floor division, floor(a/b) |
►Ctvm::tir::BinaryOpNode< SubNode > | |
Ctvm::tir::SubNode |
|
►Ctvm::tir::CmpOpNode< LTNode > | |
Ctvm::tir::LTNode | < b |
►Ctvm::tir::CmpOpNode< LENode > | |
Ctvm::tir::LENode | <= b |
►Ctvm::tir::CmpOpNode< NENode > | |
Ctvm::tir::NENode | != b |
►Ctvm::tir::CmpOpNode< GTNode > | |
Ctvm::tir::GTNode | > b |
►Ctvm::tir::CmpOpNode< EQNode > | |
Ctvm::tir::EQNode | == b |
►Ctvm::tir::CmpOpNode< GENode > | |
Ctvm::tir::GENode | >= b |
Ctvm::FloatImmNode | Constant floating point literals in the program |
Ctvm::IntImmNode | Constant integer literals in the program |
►Ctvm::arith::IterMapExprNode | Base class of all iter map expressions |
Ctvm::arith::IterSplitExprNode | Split of an iterator |
Ctvm::arith::IterSumExprNode | Fuse multiple iterators by summing them with scaling |
Ctvm::tir::AndNode | && b |
Ctvm::tir::AnyNode | Any shape |
Ctvm::tir::BinaryOpNode< T > | Base template to implement binary ops |
Ctvm::tir::BroadcastNode | Create a vector where all the elements are value |
Ctvm::tir::BufferLoadNode | Load value from the high dimension buffer |
Ctvm::tir::CallNode | Call node |
Ctvm::tir::CastNode | Cast value from one data type to another |
Ctvm::tir::CmpOpNode< T > | Base template to implement comparison ops |
Ctvm::tir::LetNode | Let binding. Bind var to value then evaluate body |
Ctvm::tir::NotNode | !a |
Ctvm::tir::OrNode | || b |
Ctvm::tir::ProducerLoadNode | Load value from the result produced by the producer |
Ctvm::tir::RampNode | 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 |
Ctvm::tir::ReduceNode | Reduction operator |
Ctvm::tir::SelectNode | Return true_value if condition is true, otherwise return false_value |
Ctvm::tir::ShuffleNode | Shuffle instruction. vec = concat(vectors) result = (vec[indices[0]], vec[indices[1]] ...) |
Ctvm::tir::StringImmNode | String constants, only used in asserts |
►Ctvm::tir::VarNode | A variable node in the IR |
Ctvm::tir::SizeVarNode | A variable node represent a tensor index size, whose value must be non-negative |
►Ctvm::RelayExprNode | Base node of all non-primitive expressions |
►Ctvm::BaseFuncNode | Base node of all functions |
Ctvm::relax::ExternFuncNode | The extern function, which can represent packed function |
Ctvm::relax::FunctionNode | A Relax function |
Ctvm::relay::FunctionNode | Relay Function container |
Ctvm::tir::PrimFuncNode | Primitive functions that contains TIR statements |
Ctvm::ConstructorNode | ADT constructor. Constructors compare by pointer equality |
Ctvm::GlobalVarNode | Global variable that lives in the top-level module |
Ctvm::OpNode | Primitive Op(builtin intrinsics) |
Ctvm::relax::CallNode | Call corresponds to callable invocation. Corresponds to operation in computational graph terminology |
Ctvm::relax::IfNode | Condition expression |
►Ctvm::relax::LeafExprNode | Base type of all (non-function) leaf Exprs |
Ctvm::relax::ConstantNode | Constant tensor |
Ctvm::relax::DataTypeImmNode | Represent a data type constant |
Ctvm::relax::PrimValueNode | PrimValue |
Ctvm::relax::ShapeExprNode | A shape expression which allows users to construct a shape containing PrimExpr |
Ctvm::relax::StringImmNode | Represent a string literal constant |
►Ctvm::relax::VarNode | The variable class for all Relax bindings |
Ctvm::relax::DataflowVarNode | A sub-type of the variable node used to mark dataflow variables from normal visible "function local" bindings |
Ctvm::relax::SeqExprNode | A sequence of blocks followed by an expression |
Ctvm::relax::TupleGetItemNode | Get index-th field out of a tuple |
Ctvm::relax::TupleNode | Tuple container |
Ctvm::relay::CallNode | Call container |
Ctvm::relay::ConstantNode | Constant tensor type |
Ctvm::relay::IfNode | Container of If |
Ctvm::relay::LetNode | A binding of a sub-network |
Ctvm::relay::MatchNode | Match container node |
Ctvm::relay::RefCreateNode | |
Ctvm::relay::RefReadNode | |
Ctvm::relay::RefWriteNode | |
Ctvm::relay::TempExprNode | Base class of the temporary expression |
Ctvm::relay::TupleGetItemNode | |
Ctvm::relay::TupleNode | Tuple container |
Ctvm::relay::VarNode | Container for Var |
Ctvm::CompilationConfigNode | 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 |
Ctvm::ConstantInfoNode | |
Ctvm::ConstantMemoryPoolsNode | |
Ctvm::DiagnosticContextNode | |
Ctvm::DiagnosticNode | A compiler diagnostic message |
Ctvm::DiagnosticRendererNode | Display diagnostics in a given display format |
Ctvm::EnvFuncNode | A serializable function backed by TVM's global environment |
Ctvm::GenericFuncNode | Represents a generic function that can be specialized on a per-target basis |
►Ctvm::GlobalInfoNode | GlobalInfo are globally static object that are referred by the IR itself. Base node for all global info that can appear in the IR |
Ctvm::DummyGlobalInfoNode | A dummy global info sub-class for testing purpose |
Ctvm::VDeviceNode | A global info subclass for virtual devices |
Ctvm::relax::distributed::DeviceMeshNode | |
Ctvm::GlobalVarSupplyNode | GlobalVarSupply can be used to generate unique GlobalVars |
Ctvm::IRModuleNode | IRModule that holds functions and type definitions |
Ctvm::MemoryInfoNode | Memory information of special memory region. Use MemoryInfo as its container type |
Ctvm::NameSupplyNode | NameSupply can be used to generate unique names |
►Ctvm::ObjectPathNode | Path to an object from some root object |
Ctvm::ArrayIndexPathNode | |
Ctvm::AttributeAccessPathNode | |
Ctvm::MapValuePathNode | |
Ctvm::MissingArrayElementPathNode | |
Ctvm::MissingMapEntryPathNode | |
Ctvm::RootPathNode | |
Ctvm::UnknownAttributeAccessPathNode | |
Ctvm::ObjectPathPairNode | Pair of ObjectPath s, one for each object being tested for structural equality |
►Ctvm::PoolInfoNode | Describes a pool of memory accessible by one or more targets |
Ctvm::ConstantPoolInfoNode | |
Ctvm::WorkspacePoolInfoNode | |
Ctvm::PoolInfoPropertiesNode | Describes a pool of memory properties |
Ctvm::PrinterConfigNode | |
Ctvm::RangeNode | Range over one dimension |
Ctvm::SourceMapNode | Stores locations in frontend source that generated a node |
Ctvm::SourceNameNode | The name of a source fragment |
Ctvm::SourceNode | |
►Ctvm::SpanNode | Stores locations in frontend source that generated a node |
Ctvm::SequentialSpanNode | Store a list of spans for an expr generated from mulitple source exprs |
Ctvm::TargetKindNode | Target kind, specifies the kind of the target |
Ctvm::TargetNode | Compilation target |
Ctvm::TargetTagNode | A target tag |
►Ctvm::TypeNode | Type is the base type of all types |
►Ctvm::BaseTensorTypeNode | Base of all Tensor types This container can hold TensorType or GenericTensorType |
Ctvm::TensorTypeNode | This is the most commonly used type in relay. TensorType have a fixed dimension, data type |
Ctvm::relax::DynTensorTypeNode | |
Ctvm::FuncTypeNode | Function type |
Ctvm::GlobalTypeVarNode | A global type variable that is used for defining new types or type aliases |
Ctvm::IncompleteTypeNode | Intermediate values that is used to indicate incomplete type during type inference |
Ctvm::PointerTypeNode | Low-level raw pointer type |
Ctvm::PrimTypeNode | Primitive data types used in the low-level IR |
Ctvm::RelayRefTypeNode | Reference Type High-level Relay IR |
Ctvm::TupleTypeNode | The type of tuple values |
Ctvm::TypeCallNode | Type function application |
►Ctvm::TypeConstraintNode | Potential Constraints in a function |
Ctvm::TypeRelationNode | User defined type relation, it is an input-output relation on types |
Ctvm::TypeDataNode | TypeData container node |
Ctvm::TypeVarNode | Type parameter in functions |
Ctvm::relax::ObjectTypeNode | |
Ctvm::relax::PackedFuncTypeNode | |
Ctvm::relax::ShapeTypeNode | |
Ctvm::TypeReporterNode | Reporter that reports back to the type resolution information |
Ctvm::WorkspaceMemoryPoolsNode | |
Ctvm::arith::ConstIntBoundNode | Constant integer up and lower bound(inclusive). Useful for value bound analysis |
Ctvm::arith::IntConstraintsNode | Represent integer constrains including (integer) variables, their ranges and the relations between them (either equations or inequalities) |
Ctvm::arith::IntConstraintsTransformNode | 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} |
Ctvm::arith::IntGroupBoundsNode | 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 |
Ctvm::arith::IntSetNode | Base class of all Integer set containers. represent a set of integers in one dimension |
Ctvm::arith::IterMapResultNode | Result of DetectIterMap |
Ctvm::arith::IterMarkNode | Mark the source as an iterator in [0, extent) |
Ctvm::arith::ModularSetNode | Range of a linear integer function. Use to do specify the possible index values |
Ctvm::auto_scheduler::AccessAnalyzerNode | Static analyzer for a ComputeDAG |
Ctvm::auto_scheduler::AttachMapNode | Stores the compute_at relation between stages This stores a bi-directional mapping from stages and iter: |
Ctvm::auto_scheduler::BuildResultNode | Store the result of a build |
Ctvm::auto_scheduler::ComputeDAGNode | The auto-scheduler's computational graph and related program analyses |
►Ctvm::auto_scheduler::CostModelNode | The base class for cost model |
Ctvm::auto_scheduler::PythonBasedModelNode | A wrapper for cost model defined by python code This class will call functions defined in the python |
Ctvm::auto_scheduler::RandomModelNode | The cost model returning random value for all predictions |
Ctvm::auto_scheduler::HardwareParamsNode | The parameters of target hardware used to guide the SearchPolicy |
Ctvm::auto_scheduler::IteratorNode | An iterator of a for-loop Similar to tvm::IterVar in include/tvm/tir/expr.h |
►Ctvm::auto_scheduler::MeasureCallbackNode | Bass class of measurement callbacks |
Ctvm::auto_scheduler::PythonBasedMeasureCallbackNode | A wrapper for measure callback defined by python code This class will call functions defined in the python |
Ctvm::auto_scheduler::RecordToFileNode | Callback for logging the input and results of measurements to file |
Ctvm::auto_scheduler::MeasureInputNode | Store the input of a measurement |
Ctvm::auto_scheduler::MeasureResultNode | Store the results of a measurement |
►Ctvm::auto_scheduler::ProgramBuilderNode | ProgramBuilder that builds the programs |
Ctvm::auto_scheduler::LocalBuilderNode | LocalBuilder use local CPU cores to build programs in parallel |
Ctvm::auto_scheduler::ProgramMeasurerNode | Measurer that measures the time costs of tvm programs This class combines ProgramBuilder and ProgramRunner, and provides a simpler API |
►Ctvm::auto_scheduler::ProgramRunnerNode | ProgramRunner that runs the built programs and measure the time cost |
Ctvm::auto_scheduler::LocalRunnerNode | LocalRunner that uses local CPU/GPU to measure the time cost of programs |
Ctvm::auto_scheduler::RPCRunnerNode | 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) |
Ctvm::auto_scheduler::RecordReaderNode | Log reader to load step logs from a file |
►Ctvm::auto_scheduler::SearchCallbackNode | 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 |
Ctvm::auto_scheduler::PreloadMeasuredStatesNode | Preload measured states from a log file. This can resume the state of the search policy |
Ctvm::auto_scheduler::SearchPolicyNode | The base class of search policies |
Ctvm::auto_scheduler::SearchTaskNode | The computation information and hardware parameters for a specific schedule search task |
Ctvm::auto_scheduler::StageNode | A op stage in the compute declaration. Similar to te::Stage in include/tvm/te/schedule.h |
Ctvm::auto_scheduler::StateNode | 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 |
►Ctvm::auto_scheduler::StepNode | The base class of transformation steps. Each step has its corresponding tvm.te schedule primitives |
Ctvm::auto_scheduler::AnnotationStepNode | 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) |
Ctvm::auto_scheduler::CacheReadStepNode | Cache read step that corresponds to te::Schedule::cache_read |
Ctvm::auto_scheduler::CacheWriteStepNode | Cache write step that corresponds to te::Schedule::cache_write |
Ctvm::auto_scheduler::ComputeAtStepNode | Compute at step that corresponds to te::Stage::compute_at |
Ctvm::auto_scheduler::ComputeInlineStepNode | Compute inline step that corresponds to te::Stage::compute_inline |
Ctvm::auto_scheduler::ComputeRootStepNode | Compute root step that corresponds to te::Stage::compute_root |
Ctvm::auto_scheduler::FollowFusedSplitStepNode | Similar to FollowSplitStep, but uses split factors from multiple steps |
Ctvm::auto_scheduler::FollowSplitStepNode | Similar to SplitStepNode, but uses split factors from another step (i.e. Follow another split step) |
Ctvm::auto_scheduler::FuseStepNode | Fuse step that corresponds to te::Stage::fuse |
Ctvm::auto_scheduler::PragmaStepNode | Pragma step that corresponds to te::Stage::pragma |
Ctvm::auto_scheduler::ReorderStepNode | Reorder step that corresponds to te::Stage::reorder |
Ctvm::auto_scheduler::RfactorStepNode | Reduction factor step that corresponds to te::Schedule::rfactor |
Ctvm::auto_scheduler::SplitStepNode | Split step that corresponds to te::Stage::split with additional support of multiple-level of factors |
Ctvm::auto_scheduler::StorageAlignStepNode | Storage align step that corresponds to te::Stage::storage_align |
Ctvm::auto_scheduler::TuningOptionsNode | Tuning and measurement options |
Ctvm::instrument::PassInstrumentNode | PassInstrumentNode forms an instrument implementation. It provides API for users to register callbacks at different instrumentation points |
►Ctvm::meta_schedule::ArgInfoNode | The argument information |
Ctvm::meta_schedule::TensorInfoNode | The tensor argument information |
Ctvm::meta_schedule::BuilderInputNode | The builder's input, containing an IRModule and the target |
►Ctvm::meta_schedule::BuilderNode | The abstract builder interface |
Ctvm::meta_schedule::PyBuilderNode | An abstract builder with customized build method on the python-side |
Ctvm::meta_schedule::BuilderResultNode | The builder's output, containing the artifact path or error message if any |
►Ctvm::meta_schedule::CostModelNode | Cost model |
Ctvm::meta_schedule::PyCostModelNode | The cost model with customized methods on the python-side |
►Ctvm::meta_schedule::DatabaseNode | |
Ctvm::meta_schedule::PyDatabaseNode | The database with customized methods on the python-side |
Ctvm::meta_schedule::ExtractedTaskNode | A tuning task extracted from the high-level IR |
►Ctvm::meta_schedule::FeatureExtractorNode | Extractor for features from measure candidates for use in cost model |
Ctvm::meta_schedule::PyFeatureExtractorNode | The feature extractor with customized methods on the python-side |
►Ctvm::meta_schedule::MeasureCallbackNode | Rules to apply after measure results is available |
Ctvm::meta_schedule::PyMeasureCallbackNode | The measure callback with customized methods on the python-side |
Ctvm::meta_schedule::MeasureCandidateNode | The schedule (with input shapes) to be measured |
►Ctvm::meta_schedule::MutatorNode | Mutator is designed to mutate the trace to explore the design space |
Ctvm::meta_schedule::PyMutatorNode | The mutator with customized methods on the python-side |
►Ctvm::meta_schedule::PostprocNode | Rules to apply a postprocessor to a schedule |
Ctvm::meta_schedule::PyPostprocNode | The postprocessor with customized methods on the python-side |
Ctvm::meta_schedule::ProfilerNode | A generic profiler |
Ctvm::meta_schedule::RunnerFutureNode | A class to asynchronously fetch runner's output |
Ctvm::meta_schedule::RunnerInputNode | Runner's input containing path of artifact, type of device and argument info |
►Ctvm::meta_schedule::RunnerNode | The abstract runner interface |
Ctvm::meta_schedule::PyRunnerNode | An abstract runner with customized build method on the python-side |
Ctvm::meta_schedule::RunnerResultNode | Runner's output containing measurement result of MeasureCandidate or error msg if any |
►Ctvm::meta_schedule::ScheduleRuleNode | Rules to modify a block in a schedule |
Ctvm::meta_schedule::PyScheduleRuleNode | The schedule rule with customized methods on the python-side |
►Ctvm::meta_schedule::SearchStrategyNode | The search strategy for measure candidates generation |
Ctvm::meta_schedule::PySearchStrategyNode | The python side customizable class for measure candidate generation |
►Ctvm::meta_schedule::SpaceGeneratorNode | The abstract class for design space generation |
Ctvm::meta_schedule::PySpaceGeneratorNode | The design space generator with customized methods on the python-side |
Ctvm::meta_schedule::TaskRecordNode | |
►Ctvm::meta_schedule::TaskSchedulerNode | The abstract interface of task schedulers |
Ctvm::meta_schedule::PyTaskSchedulerNode | The task scheduler with customized methods on the python-side |
Ctvm::meta_schedule::TuneContextNode | The auto tuning context |
Ctvm::meta_schedule::TuningRecordNode | The class of tuning records |
Ctvm::meta_schedule::WorkloadNode | A workload, i.e. an IRModule and its structural hash |
►Ctvm::relax::BindingBlockNode | |
Ctvm::relax::DataflowBlockNode | |
►Ctvm::relax::BindingNode | The base class of a variable binding in Relax |
Ctvm::relax::MatchCastNode | Runtime-match the value to the struct info |
Ctvm::relax::VarBindingNode | |
Ctvm::relax::BlockBuilderNode | A builder to build Relax binding blocks |
Ctvm::relax::ChoiceNode | Choice manages a set of keys for transformation and constraint functions |
►Ctvm::relax::DFConstraintNode | Additional constraints on the graph |
Ctvm::relax::SameShapeConstraintNode | A pattern that asserting multiple root patterns have the same shape |
►Ctvm::relax::DFPatternNode | Base type of all dataflow patterns |
Ctvm::relax::AndPatternNode | Match a conjunction of other patterns |
Ctvm::relax::AttrPatternNode | A pattern that asserting a root pattern has certain attributes |
Ctvm::relax::CallPatternNode | A pattern to match a callable node in Relax |
Ctvm::relax::ConstantPatternNode | A Pattern to Match a Relax Constant |
Ctvm::relax::DataTypePatternNode | A pattern that asserting a root pattern has a certain data type |
Ctvm::relax::ExprPatternNode | Pattern for Relax Expression |
Ctvm::relax::ExternFuncPatternNode | A pattern of external function |
Ctvm::relax::FunctionPatternNode | A pattern to match a Relax Function |
Ctvm::relax::NotPatternNode | Pattern for rejecting a certain pattern |
Ctvm::relax::OrPatternNode | Match a disjunction of other patterns |
Ctvm::relax::PrimArrPatternNode | A pattern to match an array of PrimExpr |
Ctvm::relax::ShapePatternNode | A pattern that asserting a root pattern has a certain shape |
Ctvm::relax::StructInfoPatternNode | Pattern for matching a certain struct info |
Ctvm::relax::TupleGetItemPatternNode | A pattern to match n'th indexing to a tuple |
Ctvm::relax::TuplePatternNode | Pattern to match a tuple of ordered expressions |
Ctvm::relax::TypePatternNode | Pattern for matching a certain type |
Ctvm::relax::UnorderedTuplePatternNode | A pattern to match multiple expressions unorderedly |
►Ctvm::relax::VarPatternNode | A Pattern to Match a Relax Variable |
Ctvm::relax::DataflowVarPatternNode | A Pattern to Match a Relax Dataflow Variable |
Ctvm::relax::GlobalVarPatternNode | A Pattern to Match a Relax Global Variable |
Ctvm::relax::WildcardPatternNode | Wildcard Pattern is a pattern that can match anything |
Ctvm::relax::DatabaseNode | |
Ctvm::relax::DataflowBlockRewriteNode | Statement rewriter for relax.DataflowBlock |
Ctvm::relax::ExecBuilderNode | A builder provides api to build VM executable with instructions |
Ctvm::relax::IdNode | The unique identifier of variables |
Ctvm::relax::KnobNode | Knob manages a set of valid choices for an optimization |
Ctvm::relax::MatchResultNode | |
Ctvm::relax::PatternContextNode | A context to manage the graph-level pattern matching |
Ctvm::relax::PatternSeqNode | A sequence of DFPatterns that the previous DFPattern is connected to the next one |
►Ctvm::relax::StructInfoNode | Base type of all structure information |
Ctvm::relax::FuncStructInfoNode | Structure information about function |
Ctvm::relax::ObjectStructInfoNode | Opaque object |
Ctvm::relax::PrimStructInfoNode | Primitive value |
Ctvm::relax::ShapeStructInfoNode | StructInfo of shape value |
Ctvm::relax::TensorStructInfoNode | StructInfo of Tensor |
Ctvm::relax::TupleStructInfoNode | StructInfo of Tuple |
Ctvm::relax::distributed::DTensorStructInfoNode | StructInfo of DTensor (Distributed Tensor) |
Ctvm::relax::TraceNode | Trace manages history of optimization decisions |
Ctvm::relax::TuningRecordNode | The class of tuning records |
Ctvm::relax::distributed::PlacementNode | Describes how data is distributed in each dimension of the device mesh |
►Ctvm::relax::distributed::PlacementSpecNode | Describes how data is distributed in one dimension of the device mesh |
Ctvm::relax::distributed::ShardingNode | |
Ctvm::relax::transform::FusionPatternNode | The pattern object used as the input of FuseOpsByPattern. For bindings to be fused, it needs to be matched with pattern and the check function needs to return true |
Ctvm::relax::transform::PatternCheckContextNode | The input of FusionPattern::check |
Ctvm::relay::ClauseNode | Clause container node |
Ctvm::relay::ConstructorValueObj | |
Ctvm::relay::DFPatternCallbackNode | Base type of all dataflow pattern callbacks |
►Ctvm::relay::DFPatternNode | Base type of all dataflow patterns |
Ctvm::relay::AltPatternNode | Pattern for Alternate Expressions |
Ctvm::relay::AttrPatternNode | Pattern for Attributes |
Ctvm::relay::CallPatternNode | CallPattern container |
Ctvm::relay::ConstantPatternNode | Container for Constant |
Ctvm::relay::DataTypePatternNode | Pattern for Types |
Ctvm::relay::DominatorPatternNode | 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 |
Ctvm::relay::ExprPatternNode | Pattern for Relay Expression |
Ctvm::relay::FunctionPatternNode | Relay Function container |
Ctvm::relay::IfPatternNode | |
Ctvm::relay::LetPatternNode | A binding of a sub-network |
Ctvm::relay::ShapePatternNode | Pattern for Shapes |
Ctvm::relay::TupleGetItemPatternNode | |
Ctvm::relay::TuplePatternNode | Tuple container |
Ctvm::relay::TypePatternNode | Pattern for Types |
Ctvm::relay::VarPatternNode | Container for Var |
Ctvm::relay::WildcardPatternNode | Wildcard Pattern |
Ctvm::relay::ExecutorNode | Executor information |
Ctvm::relay::IdNode | The unique identifier of variables |
Ctvm::relay::OpImplementationNode | Operator implementation that includes compute and schedule function |
Ctvm::relay::OpSpecializationNode | Specialized implementations for operators under certain conditions |
Ctvm::relay::OpStrategyNode | Operator strategy to choose implementation |
Ctvm::relay::RecClosureObj | The container type of RecClosure |
Ctvm::relay::RefValueObj | |
►Ctvm::relay::RelayNode | This is the base node container of all relay structures |
►Ctvm::relay::PatternNode | Base type for declaring relay pattern |
Ctvm::relay::PatternConstructorNode | PatternVar container node |
Ctvm::relay::PatternTupleNode | PatternVar container node |
Ctvm::relay::PatternVarNode | PatternVar container node |
Ctvm::relay::PatternWildcardNode | PatternWildcard container node |
Ctvm::relay::RuntimeNode | Runtime information |
Ctvm::runtime::ADTObj | An object representing a structure or enumeration |
Ctvm::runtime::ArrayNode | Array node content in array |
Ctvm::runtime::BoxNode< Prim > | |
►Ctvm::runtime::ClosureObj | An object representing a closure. This object is used by both the Relay VM and interpreter |
Ctvm::relay::InterpreterClosureObj | The container type of Closures used by the interpreter |
Ctvm::runtime::relax_vm::VMClosureObj | An object representing a vm closure |
Ctvm::runtime::vm::VMClosureObj | An object representing a vm closure |
Ctvm::runtime::DRefObj | An object that exists on all workers |
►Ctvm::runtime::MapNode | Shared content of all specializations of hash map |
Ctvm::runtime::DenseMapNode | A specialization of hash map that implements the idea of array-based hash map. Another reference implementation can be found [1] |
Ctvm::runtime::SmallMapNode | A specialization of small-sized hash map |
►Ctvm::runtime::ModuleNode | Base container of module |
Ctvm::runtime::relax_vm::Executable | The executable emitted by the VM compiler |
Ctvm::runtime::relax_vm::VirtualMachine | The virtual machine |
Ctvm::runtime::vm::Executable | The executable emitted by the VM compiler |
Ctvm::runtime::vm::VirtualMachine | The virtual machine |
Ctvm::runtime::NDArray::Container | Object container class that backs NDArray |
►Ctvm::runtime::PackedFuncObj | Object container class that backs PackedFunc |
Ctvm::runtime::PackedFuncSubObj< TCallable > | Derived object class for constructing PackedFuncObj |
Ctvm::runtime::SessionObj | A Disco interactive session. It allows users to interact with the Disco command queue with various PackedFunc calling convention |
►Ctvm::runtime::ShapeTupleObj | An object representing a shape tuple |
Ctvm::runtime::ShapeTupleObj::FromStd | An object representing shape tuple moved from std::vector |
►Ctvm::runtime::StringObj | An object representing string. It's POD type |
Ctvm::runtime::StringObj::FromStd | An object representing string moved from std::string |
Ctvm::runtime::TimerNode | Base class for all implementations |
Ctvm::runtime::cuda_ipc::CUDAIPCMemoryObj | The CUDA IPC (interprocess communication) memory object, which internally contains data pointers to CUDA IPC memory. It is be useful for efficient all-reduce implementation |
Ctvm::runtime::memory::StorageObj | An object representing a storage allocation |
►Ctvm::runtime::metadata::MetadataBaseNode | Common base class for all Metadata |
Ctvm::runtime::metadata::ConstantInfoMetadataNode | |
Ctvm::runtime::metadata::MetadataArrayNode | Container for arrays in the metadata |
Ctvm::runtime::metadata::MetadataNode | |
Ctvm::runtime::metadata::TensorInfoNode | |
Ctvm::runtime::profiling::CountNode | |
Ctvm::runtime::profiling::DeviceWrapperNode | Wrapper for Device because Device is not passable across the PackedFunc interface |
Ctvm::runtime::profiling::DurationNode | |
Ctvm::runtime::profiling::MetricCollectorNode | Interface for user defined profiling metric collection |
Ctvm::runtime::profiling::PercentNode | |
Ctvm::runtime::profiling::RatioNode | |
Ctvm::runtime::profiling::ReportNode | Data collected from a profiling run. Includes per-call metrics and per-device metrics |
Ctvm::runtime::relax_vm::VMExtensionNode | Represent a VM extension. A VM extension allows the user to extend the VM with target specific functionalities. The VM holds the reference of the extensions to ensure the extensions have the same lifetime as the VM |
►Ctvm::script::ir_builder::IRBuilderFrameNode | |
Ctvm::script::ir_builder::ir::IRModuleFrameNode | A frame that represents the IRModule frame with functions and global variables |
►Ctvm::script::ir_builder::relax::RelaxFrameNode | The base ir_builder frame for the relax dialect |
Ctvm::script::ir_builder::relax::BlockFrameNode | The ir_builder frame for relax binding blocks |
Ctvm::script::ir_builder::relax::IfFrameNode | A frame that represents if statement |
►Ctvm::script::ir_builder::relax::SeqExprFrameNode | The base ir_builder frame for frames with SeqExpr i.e. Functions, If branches |
Ctvm::script::ir_builder::relax::ElseFrameNode | A frame that represents else |
Ctvm::script::ir_builder::relax::FunctionFrameNode | The ir_builder frame for the relax function |
Ctvm::script::ir_builder::relax::ThenFrameNode | A frame that represents then |
►Ctvm::script::ir_builder::tir::TIRFrameNode | A base frame that represents the TIR fame with body of statements |
Ctvm::script::ir_builder::tir::AllocateConstFrameNode | A frame represents the allocate constant |
Ctvm::script::ir_builder::tir::AllocateFrameNode | A frame represents the allocate |
Ctvm::script::ir_builder::tir::AssertFrameNode | A frame that represents the assert statement. Proceeds if the condition is true, otherwise aborts with the message |
Ctvm::script::ir_builder::tir::AttrFrameNode | A frame that represents attribute node |
Ctvm::script::ir_builder::tir::BlockFrameNode | A frame that represents the block |
Ctvm::script::ir_builder::tir::BlockInitFrameNode | A frame that represents the block initialization statment |
Ctvm::script::ir_builder::tir::DeclBufferFrameNode | |
Ctvm::script::ir_builder::tir::ElseFrameNode | A frame that represents else |
Ctvm::script::ir_builder::tir::ForFrameNode | A frame that represents the for loop |
Ctvm::script::ir_builder::tir::IfFrameNode | A frame that represents if statement |
Ctvm::script::ir_builder::tir::LaunchThreadFrameNode | The LaunchThreadFrameNode |
Ctvm::script::ir_builder::tir::LetFrameNode | A frame represents the let binding expression, which binds a var |
Ctvm::script::ir_builder::tir::PrimFuncFrameNode | A frame that represents the PrimFunc containing TIR statements |
Ctvm::script::ir_builder::tir::RealizeFrameNode | A frame that represents realization |
Ctvm::script::ir_builder::tir::ThenFrameNode | A frame that represents then |
Ctvm::script::ir_builder::tir::WhileFrameNode | A frame that represents while loop |
Ctvm::script::ir_builder::IRBuilderNode | 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 |
►Ctvm::script::printer::DocNode | The base class of all Doc |
►Ctvm::script::printer::ExprDocNode | The base class of expression doc |
Ctvm::script::printer::AttrAccessDocNode | Doc that represents attribute access on another expression |
Ctvm::script::printer::CallDocNode | Doc that represents function call |
Ctvm::script::printer::DictDocNode | Doc that represents dictionary literal |
Ctvm::script::printer::IdDocNode | Doc that represents identifier |
Ctvm::script::printer::IndexDocNode | Doc that represents index access on another expression |
Ctvm::script::printer::LambdaDocNode | Doc that represents anonymous function |
Ctvm::script::printer::ListDocNode | Doc that represents list literal |
Ctvm::script::printer::LiteralDocNode | Doc that represents literal value |
Ctvm::script::printer::OperationDocNode | Doc that represents operation |
Ctvm::script::printer::TupleDocNode | Doc that represents tuple literal |
Ctvm::script::printer::SliceDocNode | Doc that represents slice in Index expression |
Ctvm::script::printer::StmtBlockDocNode | The container doc that holds a list of StmtDoc |
►Ctvm::script::printer::StmtDocNode | The base class of statement doc |
Ctvm::script::printer::AssertDocNode | Doc that represents assert statement |
Ctvm::script::printer::AssignDocNode | Doc that represents assign statement |
Ctvm::script::printer::ClassDocNode | Doc that represents class definition |
Ctvm::script::printer::CommentDocNode | Doc that represents comment |
Ctvm::script::printer::DocStringDocNode | Doc that represents docstring |
Ctvm::script::printer::ExprStmtDocNode | Doc that represents an expression as statement |
Ctvm::script::printer::ForDocNode | Doc that represents for statement |
Ctvm::script::printer::FunctionDocNode | Doc that represents function definition |
Ctvm::script::printer::IfDocNode | Doc that represent if-then-else statement |
Ctvm::script::printer::ReturnDocNode | Doc that represents return statement |
Ctvm::script::printer::ScopeDocNode | Doc that represents special scopes |
Ctvm::script::printer::WhileDocNode | Doc that represents while statement |
Ctvm::script::printer::FrameNode | |
Ctvm::script::printer::IRDocsifierNode | IRDocsifier is the top-level interface in the IR->Doc process |
Ctvm::te::IterVarAttrNode | Node container for IterVar attr |
►Ctvm::te::IterVarRelationNode | Base node of iteration var |
Ctvm::te::FuseNode | Fuse two domains into one domain |
Ctvm::te::RebaseNode | 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 |
Ctvm::te::SingletonNode | Singleton iterator [0, 1) |
Ctvm::te::SplitNode | Split the parent domain into product of outer and iter |
Ctvm::te::TransformNode | Transform iterator according to some arbitrary expression |
►Ctvm::te::OperationNode | Base class of all operation nodes |
►Ctvm::te::BaseComputeOpNode | 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) |
Ctvm::te::ComputeOpNode | A Compute op that compute a tensor on certain domain |
Ctvm::te::TensorComputeOpNode | A TenorCompute op that compute a tensor with an tensor intrinsic |
Ctvm::te::ExternOpNode | External computation that cannot be splitted |
Ctvm::te::HybridOpNode | A computation operator that generated by hybrid script |
Ctvm::te::PlaceholderOpNode | A placeholder op represents an input placeholder |
Ctvm::te::ScanOpNode | Symbolic scan |
Ctvm::te::ScheduleNode | Node container for schedule |
Ctvm::te::SpecializedConditionNode | Container for specialization conditions |
Ctvm::te::StageNode | Stage |
Ctvm::te::TensorIntrinCallNode | |
Ctvm::te::TensorIntrinNode | Node to represent a Tensor intrinsic operator |
Ctvm::tir::BijectiveLayoutNode | |
Ctvm::tir::BlockDependenceInfoNode | An object that helps build and query block level dependences using the 2 core objects BlockScope and StmtSRef |
Ctvm::tir::BlockRVNode | A random variable that evaluates to a TensorIR block |
Ctvm::tir::BlockScopeNode | 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 |
Ctvm::tir::BufferNode | Node to represent a buffer |
Ctvm::tir::BufferRegionNode | Representing the region of multi-dimensional buffer access |
Ctvm::tir::CommReducerNode | A commutative reducer node to represent a commutative binary operator with identity element |
►Ctvm::tir::DataProducerNode | Base node for data producers |
Ctvm::te::TensorNode | Node to represent a tensor |
Ctvm::tir::DependencyNode | 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 |
Ctvm::tir::IndexMapNode | Defines a mapping between two representations of indices into a buffer |
Ctvm::tir::InstructionKindNode | 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 |
Ctvm::tir::InstructionNode | Schedule instructions each corresponds to a schedule primitive |
Ctvm::tir::IterVarNode | An iteration variable representing an iteration over a one dimensional interval |
Ctvm::tir::LayoutNode | 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 |
Ctvm::tir::LoopRVNode | A random variable that evaluates to a TensorIR for loop |
Ctvm::tir::MatchBufferRegionNode | 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) |
Ctvm::tir::ScheduleNode | The user-facing schedule class |
Ctvm::tir::ScheduleStateNode | The state of scheduling, which exposes a Replace method as the primary interface for all the scheduling primitives to manipulate the TensorIR |
►Ctvm::tir::StmtNode | Base node of all statements |
Ctvm::tir::AllocateConstNode | Allocate a buffer that can be used in body |
Ctvm::tir::AllocateNode | Allocate a buffer that can be used in body |
Ctvm::tir::AssertStmtNode | Assert condition, if an error occurs, return the error message |
Ctvm::tir::AttrStmtNode | Define certain auxiliary attribute for the body to be a symbolic value. This provide auxiliary information for IR passes that transforms body |
Ctvm::tir::BlockNode | A block is a basic schedule unit in TIR |
Ctvm::tir::BlockRealizeNode | A block realization node represents execution of the block at the binding values |
Ctvm::tir::BufferRealizeNode | 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 |
Ctvm::tir::BufferStoreNode | Store value to the high dimension buffer |
Ctvm::tir::DeclBufferNode | Declare a buffer that can be used in the body |
Ctvm::tir::EvaluateNode | Evaluates an expression. This is mostly used for putting a Call node into Stmt |
Ctvm::tir::ForNode | A for loop, with possible type annotations |
Ctvm::tir::IfThenElseNode | IfThenElse statement |
Ctvm::tir::LetStmtNode | Let binding, bind var to value, then run body |
Ctvm::tir::PrefetchNode | A prefetch hint for a buffer |
Ctvm::tir::ProducerRealizeNode | 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 |
Ctvm::tir::ProducerStoreNode | Store value into mult-dimensional array that will be read by the consumer of the producer |
Ctvm::tir::SeqStmtNode | The container of seq statement. Represent a sequence of statements |
Ctvm::tir::WhileNode | A While loop |
Ctvm::tir::StmtSRefNode | An object that refers to schedulable elements (block/for-loop) in TensorIR, aka "sref" |
Ctvm::tir::TensorIntrinNode | Tensor intrinsics for tensorization |
Ctvm::tir::TraceNode | An execution trace of a scheduling program |
Ctvm::tir::usmp::AllocatedPoolInfoNode | This object contains information post-allocation for PoolInfo objects |
Ctvm::tir::usmp::BufferInfoAnalysisNode | 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 |
Ctvm::tir::usmp::BufferInfoNode | Describes an abstract memory buffer that will get allocated inside a pool. The actual memory buffer in represented by PoolAllocationNode after static memory planning |
Ctvm::tir::usmp::PoolAllocationNode | The pool allocation produced after the USMP algorithm |
Ctvm::transform::PassContextNode | PassContextNode contains the information that a pass can rely on, such as analysis results |
Ctvm::transform::PassInfoNode | Meta data that will be used to help optimization and analysis |
►Ctvm::transform::PassNode | 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/Relax nodes |
Ctvm::transform::SequentialNode | The SequentialNode contains a set of passes that transform Relay/Relax programs from one AST to another semantically equivalent one |
Ctvm::runtime::ObjectEqual | String-aware ObjectRef hash functor |
Ctvm::runtime::ObjectHash | String-aware ObjectRef equal functor |
Ctvm::runtime::ObjectPtr< T > | A custom smart pointer for Object |
Ctvm::runtime::ObjectPtr< tvm::AttrFieldInfoNode > | |
Ctvm::runtime::ObjectPtr< tvm::runtime::Object > | |
Ctvm::runtime::ObjectPtr< tvm::runtime::relax_vm::Executable > | |
Ctvm::runtime::ObjectPtr< tvm::runtime::vm::Executable > | |
Ctvm::runtime::ObjectPtrEqual | ObjectRef equal functor |
Ctvm::runtime::ObjectPtrHash | ObjectRef hash functor |
►Ctvm::runtime::ObjectRef | Base class of all object reference |
Ctvm::runtime::Array< tvm::relay::Var > | |
Ctvm::runtime::Array< tvm::meta_schedule::ArgInfo > | |
Ctvm::runtime::Array< tvm::arith::IterSumExpr > | |
Ctvm::runtime::Array< tvm::ObjectPath > | |
Ctvm::runtime::Array< tvm::tir::BufferRegion > | |
Ctvm::runtime::Array< tvm::tir::IterVar > | |
Ctvm::runtime::Array< tvm::meta_schedule::TaskRecord > | |
Ctvm::runtime::Array< tvm::Integer > | |
Ctvm::runtime::Array< tvm::ConstantInfo > | |
Ctvm::runtime::Array< tvm::Range > | |
Ctvm::runtime::Array< tvm::tir::MatchBufferRegion > | |
Ctvm::runtime::Array< tvm::relax::Binding > | |
Ctvm::runtime::Array< tvm::meta_schedule::MeasureCallback > | |
Ctvm::runtime::Array< tvm::runtime::ObjectRef > | |
Ctvm::runtime::Array< tvm::TypeConstraint > | |
Ctvm::runtime::Array< tvm::transform::Pass > | |
Ctvm::runtime::Array< tvm::script::printer::StmtDoc > | |
Ctvm::runtime::Array< tvm::relax::BindingBlock > | |
Ctvm::runtime::Array< tvm::relax::DFPattern > | |
Ctvm::runtime::Array< tvm::script::printer::ExprDoc > | |
Ctvm::runtime::Array< tvm::te::Tensor > | |
Ctvm::runtime::Array< tvm::te::Operation > | |
Ctvm::runtime::Array< tvm::te::Stage > | |
Ctvm::runtime::Array< tvm::runtime::Optional< tvm::Integer > > | |
Ctvm::runtime::Array< tvm::TensorAffineType > | |
Ctvm::runtime::Array< tvm::runtime::Array< tvm::PrimExpr > > | |
Ctvm::runtime::Array< tvm::relax::StructInfo > | |
Ctvm::runtime::Array< tvm::PoolInfo > | |
Ctvm::runtime::Array< tvm::AttrFieldInfo > | |
Ctvm::runtime::Array< tvm::ConstantPoolInfo > | |
Ctvm::runtime::Array< Range > | |
Ctvm::runtime::Array< Region > | |
Ctvm::runtime::Array< tvm::runtime::Map< tvm::runtime::String, tvm::runtime::ObjectRef > > | |
Ctvm::runtime::Array< tvm::relax::distributed::PlacementSpec > | |
Ctvm::runtime::Array< tvm::relay::OpSpecialization > | |
Ctvm::runtime::Array< tvm::auto_scheduler::Iterator > | |
Ctvm::runtime::Array< tvm::Span > | |
Ctvm::runtime::Array< tvm::Diagnostic > | |
Ctvm::runtime::Array< tvm::script::printer::AssignDoc > | |
Ctvm::runtime::Array< tvm::TypeVar > | |
Ctvm::runtime::Array< tvm::tir::Instruction > | |
Ctvm::runtime::Array< tvm::te::IterVarRelation > | |
Ctvm::runtime::Array< tvm::script::printer::Frame > | |
Ctvm::runtime::Array< tvm::runtime::Array< tvm::Integer > > | |
Ctvm::runtime::Array< tvm::relay::Clause > | |
Ctvm::runtime::Array< tvm::relay::DFPattern > | |
Ctvm::runtime::Array< tvm::relax::Var > | |
Ctvm::runtime::Array< tvm::auto_scheduler::Stage > | |
Ctvm::runtime::Array< tvm::PrimExpr > | |
Ctvm::runtime::Array< tvm::script::printer::IdDoc > | |
Ctvm::runtime::Array< tvm::relay::OpImplementation > | |
Ctvm::runtime::Array< tvm::FloatImm > | |
Ctvm::runtime::Array< tvm::tir::Var > | |
Ctvm::runtime::Array< tvm::script::printer::Doc > | |
Ctvm::runtime::Array< tvm::IRModule > | |
Ctvm::runtime::Array< tvm::Constructor > | |
Ctvm::runtime::Array< tvm::te::Schedule > | |
Ctvm::runtime::Array< tvm::relax::Knob > | |
Ctvm::runtime::Array< tvm::Type > | |
Ctvm::runtime::Array< tvm::IntImm > | |
Ctvm::runtime::Array< tvm::instrument::PassInstrument > | |
Ctvm::runtime::Array< tvm::tir::IndexMap > | |
Ctvm::runtime::Array< tvm::script::ir_builder::IRBuilderFrame > | |
Ctvm::runtime::Array< tvm::tir::Buffer > | |
Ctvm::runtime::Array< tvm::RelayExpr > | |
Ctvm::runtime::Array< tvm::tir::Stmt > | |
Ctvm::runtime::Array< tvm::relay::Pattern > | |
Ctvm::runtime::Array< tvm::auto_scheduler::Step > | |
Ctvm::runtime::Array< tvm::arith::IterSplitExpr > | |
Ctvm::runtime::Array< tvm::runtime::String > | |
Ctvm::runtime::Array< tvm::Target > | |
Ctvm::runtime::Map< tvm::relay::Var, tvm::runtime::ObjectRef > | |
Ctvm::runtime::Map< tvm::GlobalVar, tvm::BaseFunc > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::runtime::Map< tvm::runtime::String, tvm::runtime::ObjectRef > > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::relax::DFPattern > | |
Ctvm::runtime::Map< tvm::tir::Var, tvm::PrimExpr > | |
Ctvm::runtime::Map< tvm::tir::Instruction, tvm::runtime::ObjectRef > | |
Ctvm::runtime::Map< tvm::Target, tvm::Integer > | |
Ctvm::runtime::Map< tvm::GlobalTypeVar, tvm::TypeData > | |
Ctvm::runtime::Map< tvm::te::Operation, tvm::te::Stage > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::RelayExpr > | |
Ctvm::runtime::Map< tvm::tir::Var, tvm::tir::Buffer > | |
Ctvm::runtime::Map< tvm::relax::Var, tvm::RelayExpr > | |
Ctvm::runtime::Map< tvm::SourceName, tvm::Source > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::runtime::Array< tvm::GlobalInfo > > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::relax::Choice > | |
Ctvm::runtime::Map< tvm::runtime::ObjectRef, tvm::runtime::String > | |
Ctvm::runtime::Map< tvm::tir::Var, tvm::tir::IterVar > | |
Ctvm::runtime::Map< tvm::relax::Var, tvm::runtime::Array< tvm::relax::Var > > | |
Ctvm::runtime::Map< tvm::tir::Var, tvm::Range > | |
Ctvm::runtime::Map< tvm::ObjectPath, tvm::runtime::String > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::GlobalTypeVar > | |
Ctvm::runtime::Map< tvm::RelayExpr, tvm::relax::Var > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::GlobalVar > | |
Ctvm::runtime::Map< tvm::runtime::String, tvm::runtime::ObjectRef > | |
Ctvm::runtime::Map< tvm::tir::usmp::BufferInfo, tvm::tir::Stmt > | |
Ctvm::runtime::Map< tvm::tir::Buffer, tvm::tir::Buffer > | |
Ctvm::runtime::Map< tvm::tir::IterVar, tvm::te::IterVarAttr > | |
Ctvm::runtime::Optional< tvm::meta_schedule::SearchStrategy > | |
Ctvm::runtime::Optional< tvm::meta_schedule::CostModel > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::meta_schedule::ScheduleRule > > | |
Ctvm::runtime::Optional< tvm::tir::IterVar > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::tir::BufferRegion > > | |
Ctvm::runtime::Optional< tvm::Integer > | |
Ctvm::runtime::Optional< tvm::runtime::Map< tvm::meta_schedule::Mutator, tvm::FloatImm > > | |
Ctvm::runtime::Optional< tvm::runtime::NDArray > | |
Ctvm::runtime::Optional< tvm::Range > | |
Ctvm::runtime::Optional< tvm::runtime::ObjectRef > | |
Ctvm::runtime::Optional< tvm::meta_schedule::SpaceGenerator > | |
Ctvm::runtime::Optional< tvm::runtime::Map< tvm::runtime::String, tvm::runtime::NDArray > > | |
Ctvm::runtime::Optional< tvm::script::printer::ExprDoc > | |
Ctvm::runtime::Optional< tvm::runtime::PackedFunc > | |
Ctvm::runtime::Optional< tvm::VDevice > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::meta_schedule::RunnerFuture > > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::meta_schedule::ArgInfo > > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::auto_scheduler::MeasureCallback > > | |
Ctvm::runtime::Optional< tvm::Bool > | |
Ctvm::runtime::Optional< tvm::relax::StructInfo > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::PrimExpr > > | |
Ctvm::runtime::Optional< tvm::relax::PrimValue > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::IntImm > > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::meta_schedule::Postproc > > | |
Ctvm::runtime::Optional< tvm::runtime::Map< tvm::runtime::String, tvm::runtime::ObjectRef > > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::tir::Stmt > > | |
Ctvm::runtime::Optional< tvm::relay::DFPattern > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::Integer > > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::meta_schedule::MeasureCandidate > > | |
Ctvm::runtime::Optional< tvm::PrimExpr > | |
Ctvm::runtime::Optional< tvm::TypedEnvFunc > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::relax::StructInfo > > | |
Ctvm::runtime::Optional< tvm::FloatImm > | |
Ctvm::runtime::Optional< tvm::IRModule > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::FloatImm > > | |
Ctvm::runtime::Optional< tvm::DiagnosticContext > | |
Ctvm::runtime::Optional< tvm::Type > | |
Ctvm::runtime::Optional< tvm::IntImm > | |
Ctvm::runtime::Optional< tvm::runtime::Array< tvm::meta_schedule::BuilderResult > > | |
Ctvm::runtime::Optional< tvm::RelayExpr > | |
Ctvm::runtime::Optional< tvm::relax::Function > | |
Ctvm::runtime::Optional< tvm::meta_schedule::Database > | |
Ctvm::runtime::Optional< tvm::runtime::Map< tvm::runtime::String, tvm::Bool > > | |
Ctvm::runtime::Optional< tvm::tir::Stmt > | |
Ctvm::runtime::Optional< tvm::Target > | |
Ctvm::runtime::Optional< tvm::runtime::String > | |
Ctvm::runtime::Variant< tvm::runtime::Box, tvm::runtime::Array< tvm::runtime::Box > > | |
►Ctvm::AffineType | Managed reference to AffineTypeNode |
Ctvm::TensorAffineType | Managed reference to AffineTypes |
Ctvm::TupleAffineType | Managed reference to TupleAffineTypes |
Ctvm::AttrFieldInfo | AttrFieldInfo |
►Ctvm::Attrs | Managed reference to BaseAttrsNode |
Ctvm::DictAttrs | Managed reference to DictAttrsNode |
►Ctvm::BaseExpr | Managed reference to BaseExprNode |
►Ctvm::PrimExpr | Reference to PrimExprNode |
Ctvm::FloatImm | Managed reference class to FloatImmNode |
►Ctvm::IntImm | Managed reference class to IntImmNode |
Ctvm::Bool | Boolean constant |
Ctvm::Integer | Container of constant int that adds more constructors |
►Ctvm::arith::IterMapExpr | Managed reference to IterMapExprNode |
Ctvm::arith::IterSplitExpr | Managed reference to IterSplitExprNode |
Ctvm::arith::IterSumExpr | Managed reference to IterSumExprNode |
Ctvm::tir::Add | Managed reference to AddNode |
Ctvm::tir::And | Managed reference to AndNode |
Ctvm::tir::Any | Managed reference to AnyNode |
Ctvm::tir::Broadcast | Managed reference to BroadcastNode |
Ctvm::tir::BufferLoad | Managed reference to BufferLoadNode |
Ctvm::tir::Call | Managed reference to CallNode |
Ctvm::tir::Cast | Managed reference to CastNode |
Ctvm::tir::Div | Managed reference to DivNode |
Ctvm::tir::EQ | Managed reference to EQNode |
Ctvm::tir::FloorDiv | Managed reference to FloorDivNode |
Ctvm::tir::FloorMod | Managed reference to FloorModNode |
Ctvm::tir::GE | Managed reference to GENode |
Ctvm::tir::GT | Managed reference to GTNode |
Ctvm::tir::LE | Managed reference to LENode |
Ctvm::tir::LT | Managed reference to LTNode |
Ctvm::tir::Let | Managed reference to LetNode |
Ctvm::tir::Max | Managed reference to MaxNode |
Ctvm::tir::Min | Managed reference to MinNode |
Ctvm::tir::Mod | Managed reference to ModNode |
Ctvm::tir::Mul | Managed reference to MulNode |
Ctvm::tir::NE | Managed reference to NENode |
Ctvm::tir::Not | Managed reference to NotNode |
Ctvm::tir::Or | Managed reference to OrNode |
Ctvm::tir::ProducerLoad | Managed reference to ProducerLoadNode |
Ctvm::tir::Ramp | Managed reference to RampNode |
Ctvm::tir::Reduce | Managed reference to ReduceNode |
Ctvm::tir::Select | Managed reference to SelectNode |
Ctvm::tir::Shuffle | Managed reference to ShuffleNode |
Ctvm::tir::StringImm | Managed reference to StringImmNode |
Ctvm::tir::Sub | Managed reference to SubNode |
►Ctvm::tir::Var | Named variable in TIR |
Ctvm::tir::SizeVar | Named variable represents a tensor index size |
►Ctvm::RelayExpr | Managed reference to RelayExprNode |
►Ctvm::BaseFunc | Managed reference to BaseFuncNode |
Ctvm::relax::ExternFunc | |
Ctvm::relax::Function | |
Ctvm::relay::Function | Managed reference to FunctionNode |
Ctvm::tir::PrimFunc | Managed reference to PrimFuncNode |
Ctvm::Constructor | Managed reference to ConstructorNode |
Ctvm::GlobalVar | Managed reference to GlobalVarNode |
Ctvm::Op | Managed reference class to OpNode |
Ctvm::relax::Call | |
Ctvm::relax::If | |
►Ctvm::relax::LeafExpr | Managed reference to BaseExprNode |
Ctvm::relax::Constant | |
Ctvm::relax::DataTypeImm | Managed reference to DataTypeImm |
Ctvm::relax::PrimValue | Managed reference to PrimValueNode |
Ctvm::relax::ShapeExpr | |
Ctvm::relax::StringImm | Managed reference to StringImm |
►Ctvm::relax::Var | |
Ctvm::relax::DataflowVar | |
Ctvm::relax::SeqExpr | |
Ctvm::relax::Tuple | |
Ctvm::relax::TupleGetItem | |
Ctvm::relay::Call | |
Ctvm::relay::Constant | |
Ctvm::relay::If | |
Ctvm::relay::Let | |
Ctvm::relay::Match | |
Ctvm::relay::RefCreate | |
Ctvm::relay::RefRead | |
Ctvm::relay::RefWrite | |
Ctvm::relay::TempExpr | |
Ctvm::relay::Tuple | |
Ctvm::relay::TupleGetItem | |
Ctvm::relay::Var | |
Ctvm::CompilationConfig | Managed reference class to CompilationConfig |
Ctvm::ConstantInfo | |
Ctvm::ConstantMemoryPools | |
Ctvm::Diagnostic | |
Ctvm::DiagnosticContext | |
Ctvm::DiagnosticRenderer | |
Ctvm::EnvFunc | Managed reference to EnvFuncNode |
Ctvm::GenericFunc | Generic function that can be specialized on a per-target basis |
►Ctvm::GlobalInfo | Managed reference to GlobalInfoNode |
Ctvm::DummyGlobalInfo | Managed reference to DummyGlobalInfoNode |
Ctvm::VDevice | Managed reference to VDeviceNode |
Ctvm::relax::distributed::DeviceMesh | Managed reference to a DeviceMesh |
Ctvm::GlobalVarSupply | Managed reference class to GlobalVarSupplyNode |
Ctvm::IRModule | Managed reference class to IRModuleNode |
Ctvm::MemoryInfo | Defines memory info |
Ctvm::NameSupply | Managed reference class to NameSupplyNode |
►Ctvm::ObjectPath | |
Ctvm::ArrayIndexPath | |
Ctvm::AttributeAccessPath | |
Ctvm::MapValuePath | |
Ctvm::MissingArrayElementPath | |
Ctvm::MissingMapEntryPath | |
Ctvm::RootPath | |
Ctvm::UnknownAttributeAccessPath | |
Ctvm::ObjectPathPair | |
►Ctvm::PoolInfo | Base class for WorkspacePoolInfo and ConstantPoolInfo |
Ctvm::ConstantPoolInfo | |
Ctvm::WorkspacePoolInfo | |
Ctvm::PoolInfoProperties | |
Ctvm::PrinterConfig | |
Ctvm::Range | Range container
|
Ctvm::Source | |
Ctvm::SourceMap | |
Ctvm::SourceName | The source name of a file span |
►Ctvm::Span | |
Ctvm::SequentialSpan | Reference class of SequentialSpanNode |
Ctvm::Target | Managed reference class to TargetNode |
Ctvm::TargetKind | Managed reference class to TargetKindNode |
Ctvm::TargetTag | Managed reference class to TargetTagNode |
►Ctvm::Type | Managed reference to TypeNode |
Ctvm::BaseTensorType | Managed reference to BaseTensorTypeNode |
Ctvm::FuncType | Managed reference to FuncTypeNode |
Ctvm::GlobalTypeVar | Managed reference to GlobalTypeVarNode |
Ctvm::IncompleteType | Managed reference to IncompleteTypeNode |
Ctvm::PointerType | |
Ctvm::PrimType | |
Ctvm::RelayRefType | Managed reference to RelayRefTypeNode |
Ctvm::TensorType | Managed reference to TensorTypeNode |
Ctvm::TupleType | Managed reference to TupleTypeNode |
Ctvm::TypeCall | Managed reference to TypeCallNode |
►Ctvm::TypeConstraint | Managed reference to TypeConstraintNode |
Ctvm::TypeRelation | Managed reference to TypeRelationNode |
Ctvm::TypeData | Stores all data for an Algebraic Data Type (ADT) |
Ctvm::TypeVar | Managed reference to TypeVarNode |
Ctvm::relax::DynTensorType | Managed reference to DynTensorTypeNode |
Ctvm::relax::ObjectType | |
Ctvm::relax::PackedFuncType | |
Ctvm::relax::ShapeType | |
Ctvm::TypeReporter | Container class of TypeReporter |
Ctvm::TypedEnvFunc< R(Args...)> | A typed version of EnvFunc. It is backed by a GlobalFuncNode internally |
Ctvm::VirtualDevice | Managed reference class to VirtualDeviceNode |
Ctvm::WorkspaceMemoryPools | |
Ctvm::arith::ConstIntBound | Reference class to ConstIntBoundNode |
Ctvm::arith::IntConstraints | Managed reference to IntConstraintsNode |
Ctvm::arith::IntConstraintsTransform | Managed reference to IntConstraintsTransformNode |
Ctvm::arith::IntGroupBounds | Managed reference to IntGroupBoundsNode |
Ctvm::arith::IntSet | Managed reference to IntSetNode |
Ctvm::arith::IterMapResult | Managed reference to IterMapResultNode |
Ctvm::arith::IterMark | Managed reference to IterMarkExprNode |
Ctvm::arith::ModularSet | Reference of ModularSetNode |
Ctvm::auto_scheduler::AccessAnalyzer | Managed reference to AccessAnalyzerNode |
Ctvm::auto_scheduler::AttachMap | Managed reference to AttachMapNode |
Ctvm::auto_scheduler::BuildResult | Managed reference to BuildResultNode |
Ctvm::auto_scheduler::ComputeDAG | Managed reference to ComputeDAGNode |
►Ctvm::auto_scheduler::CostModel | Managed reference to CostModelNode |
Ctvm::auto_scheduler::PythonBasedModel | Managed reference to PythonBasedModelNode |
Ctvm::auto_scheduler::RandomModel | Managed reference to RandomModelNode |
Ctvm::auto_scheduler::HardwareParams | Managed reference to HardwareParamsNode |
Ctvm::auto_scheduler::Iterator | Managed reference to IteratorNode |
►Ctvm::auto_scheduler::MeasureCallback | Managed reference to MeasureCallbackNode |
Ctvm::auto_scheduler::PythonBasedMeasureCallback | Managed reference to PythonBasedMeasureCallbackNode |
Ctvm::auto_scheduler::RecordToFile | Managed reference to RecordToFileNode |
Ctvm::auto_scheduler::MeasureInput | Managed reference to MeasureInputNode |
Ctvm::auto_scheduler::MeasureResult | Managed reference to MeasureResultNode |
►Ctvm::auto_scheduler::ProgramBuilder | Managed reference to ProgramBuilderNode |
Ctvm::auto_scheduler::LocalBuilder | Managed reference to LocalBuilderNode |
Ctvm::auto_scheduler::ProgramMeasurer | Managed reference to ProgramMeasurerNode |
►Ctvm::auto_scheduler::ProgramRunner | Managed reference to ProgramRunnerNode |
Ctvm::auto_scheduler::LocalRunner | Managed reference to LocalRunnerNode |
Ctvm::auto_scheduler::RPCRunner | Managed reference to RPCRunnerNode |
Ctvm::auto_scheduler::RecordReader | Managed reference to RecordReaderNode |
►Ctvm::auto_scheduler::SearchCallback | Managed reference to SearchCallbackNode |
Ctvm::auto_scheduler::PreloadMeasuredStates | Managed reference to PreloadMeasuredStatesNode |
Ctvm::auto_scheduler::SearchPolicy | Managed reference to SearchPolicyNode |
Ctvm::auto_scheduler::SearchTask | Managed reference to SearchTaskNode |
Ctvm::auto_scheduler::Stage | Managed reference to StageNode |
Ctvm::auto_scheduler::State | Managed reference to StateNode |
►Ctvm::auto_scheduler::Step | Managed reference to StepNode |
Ctvm::auto_scheduler::AnnotationStep | Managed reference to AnnotationStepNode |
Ctvm::auto_scheduler::CacheReadStep | Managed reference to CacheReadStepNode |
Ctvm::auto_scheduler::CacheWriteStep | Managed reference to CacheWriteStepNode |
Ctvm::auto_scheduler::ComputeAtStep | Managed reference to ComputeAtStepNode |
Ctvm::auto_scheduler::ComputeInlineStep | Managed reference to ComputeInlineStepNode |
Ctvm::auto_scheduler::ComputeRootStep | Managed reference to ComputeRootStepNode |
Ctvm::auto_scheduler::FollowFusedSplitStep | Managed reference to FollowFusedSplitStepNode |
Ctvm::auto_scheduler::FollowSplitStep | Managed reference to FollowSplitStepNode |
Ctvm::auto_scheduler::FuseStep | Managed reference to FuseStepNode |
Ctvm::auto_scheduler::PragmaStep | Managed reference to PragmaStepNode |
Ctvm::auto_scheduler::ReorderStep | Managed reference to ReorderStepNode |
Ctvm::auto_scheduler::RfactorStep | Managed reference to RfactorStepNode |
Ctvm::auto_scheduler::SplitStep | Managed reference to SplitStepNode |
Ctvm::auto_scheduler::StorageAlignStep | Managed reference to StorageAlignStepNode |
Ctvm::auto_scheduler::TuningOptions | Managed reference to TuningOptionsNode |
Ctvm::instrument::PassInstrument | Managed reference class for PassInstrumentNode |
►Ctvm::meta_schedule::ArgInfo | Managed reference to ArgInfoNode |
Ctvm::meta_schedule::TensorInfo | Managed reference to TensorInfoNode |
Ctvm::meta_schedule::Builder | Managed reference to BuilderNode |
Ctvm::meta_schedule::BuilderInput | Managed reference to BuilderInputNode |
Ctvm::meta_schedule::BuilderResult | Managed reference to BuilderResultNode |
Ctvm::meta_schedule::CostModel | Managed reference to CostModelNode |
Ctvm::meta_schedule::Database | Managed reference to DatabaseNode |
Ctvm::meta_schedule::ExtractedTask | Managed reference to ExtractedTaskNode |
Ctvm::meta_schedule::FeatureExtractor | Managed reference to FeatureExtractorNode |
Ctvm::meta_schedule::MeasureCallback | Managed reference to MeasureCallbackNode |
Ctvm::meta_schedule::MeasureCandidate | Managed reference to MeasureCandidateNode |
Ctvm::meta_schedule::Mutator | Managed reference to MutatorNode |
Ctvm::meta_schedule::Postproc | Managed reference to PostprocNode |
Ctvm::meta_schedule::Profiler | Managed reference to ProfilerNode |
Ctvm::meta_schedule::Runner | Managed reference to RunnerNode |
Ctvm::meta_schedule::RunnerFuture | Managed reference to RunnerFutureNode |
Ctvm::meta_schedule::RunnerInput | Managed reference to RunnerInputNode |
Ctvm::meta_schedule::RunnerResult | Managed reference to RunnerResultNode |
Ctvm::meta_schedule::ScheduleRule | Managed reference to ScheduleRuleNode |
Ctvm::meta_schedule::SearchStrategy | Managed reference to SearchStrategyNode |
Ctvm::meta_schedule::SpaceGenerator | Managed reference to SpaceGeneratorNode |
Ctvm::meta_schedule::TaskRecord | Managed reference to TaskRecordNode |
Ctvm::meta_schedule::TaskScheduler | Managed reference to TaskSchedulerNode |
Ctvm::meta_schedule::TuneContext | Managed reference to TuneContextNode |
Ctvm::meta_schedule::TuningRecord | The managed reference of TuningRecordNode |
Ctvm::meta_schedule::Workload | Managed reference to WorkloadNode |
►Ctvm::relax::Binding | |
Ctvm::relax::MatchCast | Managed reference to MatchCastNode |
Ctvm::relax::VarBinding | |
►Ctvm::relax::BindingBlock | |
Ctvm::relax::DataflowBlock | |
Ctvm::relax::BlockBuilder | |
Ctvm::relax::Choice | Managed reference to ChoiceNode |
►Ctvm::relax::DFConstraint | |
Ctvm::relax::SameShapeConstraint | Managed reference to SameShapePatternNode |
►Ctvm::relax::DFPattern | Managed reference to dataflow patterns |
Ctvm::relax::AndPattern | Managed reference to AndPatternNode |
Ctvm::relax::AttrPattern | Managed reference to AttrPatternNode |
Ctvm::relax::CallPattern | |
Ctvm::relax::ConstantPattern | Managed reference to a ConstantPattern |
Ctvm::relax::DataTypePattern | Managed reference to DataTypePatternNode |
Ctvm::relax::DataflowVarPattern | Managed reference to a DataflowVarPattern |
Ctvm::relax::ExprPattern | Managed reference to an ExprPattern |
Ctvm::relax::ExternFuncPattern | Managed reference to ExternFuncPatternNode |
Ctvm::relax::FunctionPattern | Managed reference to FunctionPatternNode |
Ctvm::relax::GlobalVarPattern | Managed reference to a GlobalVarPattern |
Ctvm::relax::NotPattern | Managed reference to NotPatternNode |
Ctvm::relax::OrPattern | Managed reference to OrPatternNode |
Ctvm::relax::PrimArrPattern | Managed reference to a PrimArrPattern |
Ctvm::relax::ShapePattern | Managed reference to ShapePatternNode |
Ctvm::relax::StructInfoPattern | |
Ctvm::relax::TupleGetItemPattern | Managed reference to TupleGetItemPatternNode |
Ctvm::relax::TuplePattern | Managed reference to TuplePatternNode |
Ctvm::relax::TypePattern | Managed reference to TypePatternNode |
Ctvm::relax::UnorderedTuplePattern | Managed reference to UnorderedTuplePatternNode |
Ctvm::relax::VarPattern | Managed reference to a VarPattern |
Ctvm::relax::WildcardPattern | Managed reference to WildcardPatternNode |
Ctvm::relax::Database | Managed reference to DatabaseNode |
Ctvm::relax::DataflowBlockRewrite | A statement rewriter for relax.DataflowBlock |
Ctvm::relax::ExecBuilder | |
Ctvm::relax::Id | |
Ctvm::relax::Knob | Managed reference to KnobNode |
Ctvm::relax::MatchResult | Managed reference to MatchResultNode |
Ctvm::relax::NestedMsg< T > | Container that stores possibly nested message with leaf message type T |
Ctvm::relax::PatternContext | Managed reference to a pattern context |
Ctvm::relax::PatternSeq | Managed reference to pattern sequences |
►Ctvm::relax::StructInfo | Managed reference to StructInfoNode |
Ctvm::relax::FuncStructInfo | Managed reference to FuncStructInfoNode |
Ctvm::relax::ObjectStructInfo | Managed reference to ObjectStructInfoNode |
Ctvm::relax::PrimStructInfo | Managed reference to PrimStructInfoNode |
Ctvm::relax::ShapeStructInfo | Managed reference to ShapeStructInfoNode |
Ctvm::relax::TensorStructInfo | Managed reference to TensorStructInfoNode |
Ctvm::relax::TupleStructInfo | Managed reference to TupleStructInfoNode |
Ctvm::relax::distributed::DTensorStructInfo | Managed reference to DTensorStructInfoNode |
Ctvm::relax::Trace | Managed reference to TraceNode |
Ctvm::relax::TuningRecord | The managed reference of TuningRecordNode |
Ctvm::relax::distributed::Placement | Managed reference to a Placement |
Ctvm::relax::distributed::PlacementSpec | Managed reference to PlacementSpecNode |
Ctvm::relax::transform::FusionPattern | |
Ctvm::relax::transform::PatternCheckContext | |
Ctvm::relay::Clause | |
Ctvm::relay::ConstructorValue | |
►Ctvm::relay::DFPattern | Managed reference to dataflow patterns |
Ctvm::relay::AltPattern | A pattern which matches either of two patterns |
Ctvm::relay::AttrPattern | A pattern which matches attributes in another pattern |
Ctvm::relay::CallPattern | |
Ctvm::relay::ConstantPattern | |
Ctvm::relay::DataTypePattern | A pattern which matches a type in another pattern |
Ctvm::relay::DominatorPattern | A pattern which matches a variable length dominator path |
Ctvm::relay::ExprPattern | A pattern which matches a literal expression |
Ctvm::relay::FunctionPattern | Managed reference to FunctionNode |
Ctvm::relay::IfPattern | |
Ctvm::relay::LetPattern | Let binding that binds a local var |
Ctvm::relay::ShapePattern | A pattern which matches a type in another pattern |
Ctvm::relay::TupleGetItemPattern | |
Ctvm::relay::TuplePattern | |
Ctvm::relay::TypePattern | A pattern which matches a type in another pattern |
Ctvm::relay::VarPattern | |
Ctvm::relay::WildcardPattern | A pattern which matches anything |
Ctvm::relay::DFPatternCallback | Managed reference to dataflow pattern callbacks |
Ctvm::relay::Executor | Managed reference class to ExecutorNode |
Ctvm::relay::Id | |
Ctvm::relay::OpImplementation | Operator implementation class |
Ctvm::relay::OpSpecialization | Operator specialization class |
Ctvm::relay::OpStrategy | Operator strategy class |
►Ctvm::relay::Pattern | Pattern is the base type for an ADT match pattern in Relay |
Ctvm::relay::PatternConstructor | |
Ctvm::relay::PatternTuple | |
Ctvm::relay::PatternVar | |
Ctvm::relay::PatternWildcard | |
Ctvm::relay::RecClosure | |
Ctvm::relay::RefValue | |
Ctvm::relay::Runtime | Managed reference class to RuntimeNode |
Ctvm::runtime::ADT | Reference to algebraic data type objects |
Ctvm::runtime::Array< T, typename > | Array, container representing a contiguous sequence of ObjectRefs |
Ctvm::runtime::Box< Prim > | |
►Ctvm::runtime::Closure | Reference to closure |
Ctvm::relay::InterpreterClosure | |
Ctvm::runtime::relax_vm::VMClosure | Reference to closure |
Ctvm::runtime::vm::VMClosure | Reference to closure |
Ctvm::runtime::DRef | Managed reference to DRefObj |
Ctvm::runtime::Map< K, V, typename, typename > | 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 |
Ctvm::runtime::Module | Module container of TVM |
Ctvm::runtime::NDArray | Managed NDArray. The array is backed by reference counted blocks |
Ctvm::runtime::Optional< T > | Optional container that to represent to a Nullable variant of T |
Ctvm::runtime::PackedFunc | Packed function is a type-erased function. The arguments are passed by packed format |
Ctvm::runtime::Session | Managed reference to SessionObj |
Ctvm::runtime::ShapeTuple | Reference to shape tuple objects |
Ctvm::runtime::String | Reference to string objects |
Ctvm::runtime::Timer | Timer for a specific device |
Ctvm::runtime::Variant< V > | |
Ctvm::runtime::cuda_ipc::CUDAIPCMemory | Managed reference to CUDAIPCMemoryObj |
Ctvm::runtime::memory::Storage | Reference to storage |
►Ctvm::runtime::metadata::MetadataBase | Reference class for the common MetadataBaseNode class |
Ctvm::runtime::metadata::ConstantInfoMetadata | |
Ctvm::runtime::metadata::Metadata | |
Ctvm::runtime::metadata::MetadataArray | Reference class for MetadataArray |
Ctvm::runtime::metadata::TensorInfo | |
Ctvm::runtime::profiling::DeviceWrapper | Wrapper for Device |
Ctvm::runtime::profiling::MetricCollector | Wrapper for MetricCollectorNode |
Ctvm::runtime::profiling::Report | |
Ctvm::runtime::relax_vm::VMExtension | Managed reference to VM extension |
Ctvm::script::ir_builder::IRBuilder | Managed reference to an IRBuilderNode |
►Ctvm::script::ir_builder::IRBuilderFrame | Managed reference to an IRBuilderFrameNode |
Ctvm::script::ir_builder::ir::IRModuleFrame | Managed reference to IRModuleFrameNode |
►Ctvm::script::ir_builder::relax::RelaxFrame | |
Ctvm::script::ir_builder::relax::BlockFrame | |
Ctvm::script::ir_builder::relax::IfFrame | Managed reference to IfFrameNode |
►Ctvm::script::ir_builder::relax::SeqExprFrame | |
Ctvm::script::ir_builder::relax::ElseFrame | Managed reference to ElseFrameNode |
Ctvm::script::ir_builder::relax::FunctionFrame | |
Ctvm::script::ir_builder::relax::ThenFrame | Managed reference to ThenFrameNode |
►Ctvm::script::ir_builder::tir::TIRFrame | Managed reference to TIRFrameNode |
Ctvm::script::ir_builder::tir::AllocateConstFrame | Managed reference to AllocateConstFrameNode |
Ctvm::script::ir_builder::tir::AllocateFrame | Managed reference to AllocateFrameNode |
Ctvm::script::ir_builder::tir::AssertFrame | Managed reference to AssertFrameNode |
Ctvm::script::ir_builder::tir::AttrFrame | Managed reference to AttrFrameNode |
Ctvm::script::ir_builder::tir::BlockFrame | Managed reference to BlockFrameNode |
Ctvm::script::ir_builder::tir::BlockInitFrame | Managed reference to BlockInitFrameNode |
Ctvm::script::ir_builder::tir::DeclBufferFrame | |
Ctvm::script::ir_builder::tir::ElseFrame | Managed reference to ElseFrameNode |
Ctvm::script::ir_builder::tir::ForFrame | Managed reference to ForFrameNode |
Ctvm::script::ir_builder::tir::IfFrame | Managed reference to IfFrameNode |
Ctvm::script::ir_builder::tir::LaunchThreadFrame | Managed reference to LaunchThreadFrameNode |
Ctvm::script::ir_builder::tir::LetFrame | Managed reference to LetFrameNode |
Ctvm::script::ir_builder::tir::PrimFuncFrame | Managed reference to PrimFuncFrameNode |
Ctvm::script::ir_builder::tir::RealizeFrame | Managed reference to RealizeFrameNode |
Ctvm::script::ir_builder::tir::ThenFrame | Managed reference to ThenFrameNode |
Ctvm::script::ir_builder::tir::WhileFrame | Managed reference to WhileFrameNode |
►Ctvm::script::printer::Doc | Reference type of DocNode |
►Ctvm::script::printer::ExprDoc | Reference type of ExprDocNode |
Ctvm::script::printer::AttrAccessDoc | Reference type of AttrAccessDocNode |
Ctvm::script::printer::CallDoc | Reference type of CallDocNode |
Ctvm::script::printer::DictDoc | Reference type of DictDocNode |
Ctvm::script::printer::IdDoc | Reference type of IdDocNode |
Ctvm::script::printer::IndexDoc | Reference type of IndexDocNode |
Ctvm::script::printer::LambdaDoc | Reference type of LambdaDocNode |
Ctvm::script::printer::ListDoc | Reference type of ListDocNode |
Ctvm::script::printer::LiteralDoc | Reference type of LiteralDocNode |
Ctvm::script::printer::OperationDoc | Reference type of OperationDocNode |
Ctvm::script::printer::TupleDoc | Reference type of TupleDocNode |
Ctvm::script::printer::SliceDoc | Reference type of SliceDocNode |
Ctvm::script::printer::StmtBlockDoc | Reference type of StmtBlockDocNode |
►Ctvm::script::printer::StmtDoc | Reference type of StmtDocNode |
Ctvm::script::printer::AssertDoc | Reference type of AssertDocNode |
Ctvm::script::printer::AssignDoc | Reference type of AssignDocNode |
Ctvm::script::printer::ClassDoc | Reference type of ClassDocNode |
Ctvm::script::printer::CommentDoc | Reference type of CommentDocNode |
Ctvm::script::printer::DocStringDoc | Reference type of DocStringDocNode |
Ctvm::script::printer::ExprStmtDoc | Reference type of ExprStmtDocNode |
Ctvm::script::printer::ForDoc | Reference type of ForDocNode |
Ctvm::script::printer::FunctionDoc | Reference type of FunctionDocNode |
Ctvm::script::printer::IfDoc | Reference type of IfDocNode |
Ctvm::script::printer::ReturnDoc | Reference type of ReturnDocNode |
Ctvm::script::printer::ScopeDoc | Reference type of ScopeDocNode |
Ctvm::script::printer::WhileDoc | Reference type of WhileDocNode |
Ctvm::script::printer::Frame | Reference type of FrameNode |
Ctvm::script::printer::IRDocsifier | Reference type of IRDocsifierNode |
Ctvm::te::IterVarAttr | Additional scheduable attributes about IterVar |
►Ctvm::te::IterVarRelation | The schedule relation between IterVars can be Split, Fuse |
Ctvm::te::Fuse | Managed reference to FuseNode |
Ctvm::te::Rebase | Managed reference to RebaseNode |
Ctvm::te::Singleton | Managed reference to SingletonNode |
Ctvm::te::Split | Managed reference to SplitNode |
Ctvm::te::Transform | |
►Ctvm::te::Operation | Operation that produces tensors |
Ctvm::te::ComputeOp | Managed reference to ComputeOpNode |
Ctvm::te::ExternOp | Managed reference to ExternOpNode |
Ctvm::te::HybridOp | Managed reference to HybridOpNode |
Ctvm::te::PlaceholderOp | Managed reference to PlaceholderOpNode |
Ctvm::te::ScanOp | Managed reference to ScanOpNode |
Ctvm::te::TensorComputeOp | Managed reference to TensorComputeOpNode |
Ctvm::te::Schedule | Global schedule container For operations and all the operations they depend on. The schedule per Operation is named as stage |
Ctvm::te::SpecializedCondition | Specialized condition to enable op specialization |
Ctvm::te::Stage | Stage, contains scheduling for a stage of computation |
Ctvm::te::TensorIntrin | Managed reference to TensorIntrinNode |
Ctvm::te::TensorIntrinCall | Managed reference to TensorIntrinCallNode |
Ctvm::tir::BijectiveLayout | 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) |
Ctvm::tir::BlockDependenceInfo | Managed reference to BlockDependenceInfoNode |
Ctvm::tir::BlockRV | Managed reference to BlockRVNode |
Ctvm::tir::BlockScope | Managed reference to BlockScopeNode |
Ctvm::tir::Buffer | 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 |
Ctvm::tir::BufferRegion | Managed reference to BufferRegionNode |
Ctvm::tir::CommReducer | Managed reference to CommReducerNode |
►Ctvm::tir::DataProducer | Managed reference to DataProducerNode |
Ctvm::te::Tensor | Tensor structure representing a possible input, or intermediate computation result |
Ctvm::tir::Dependency | Managed reference to DependencyNode |
Ctvm::tir::IndexMap | |
Ctvm::tir::Instruction | Managed reference to InstructionNode |
Ctvm::tir::InstructionKind | Managed reference to InstructionKindNode |
Ctvm::tir::IterVar | Iteration Variable, represents an iteration over an integer interval |
Ctvm::tir::Layout | Managed reference to LayoutNode |
Ctvm::tir::LoopRV | Managed reference to LoopRVNode |
Ctvm::tir::MatchBufferRegion | Managed reference to MatchBufferRegionNode |
Ctvm::tir::Schedule | Managed reference to ScheduleNode |
Ctvm::tir::ScheduleState | Managed reference to ScheduleStateNode |
►Ctvm::tir::Stmt | Container of all statements |
Ctvm::tir::Allocate | Managed reference to AllocateNode |
Ctvm::tir::AllocateConst | Managed reference to AllocateConstNode |
Ctvm::tir::AssertStmt | Managed reference to AssertStmtNode |
Ctvm::tir::AttrStmt | Managed reference to AttrStmtNode |
Ctvm::tir::Block | Managed reference to BlockNode |
Ctvm::tir::BlockRealize | Managed reference to BlockRealizeNode |
Ctvm::tir::BufferRealize | Managed reference to BufferRealizeNode |
Ctvm::tir::BufferStore | Managed reference to BufferStoreNode |
Ctvm::tir::DeclBuffer | Managed reference to DeclBufferNode |
Ctvm::tir::Evaluate | Managed reference to EvaluateNode |
Ctvm::tir::For | Managed reference to ForNode |
Ctvm::tir::IfThenElse | Managed reference to IfThenElseNode |
Ctvm::tir::LetStmt | Managed reference to LetStmtNode |
Ctvm::tir::Prefetch | Managed reference to PrefetchNode |
Ctvm::tir::ProducerRealize | Managed reference to ProducerRealizeNode |
Ctvm::tir::ProducerStore | Managed reference to ProducerStoreNode |
Ctvm::tir::SeqStmt | Sequence statement |
Ctvm::tir::While | Managed reference to WhileNode |
Ctvm::tir::StmtSRef | Managed reference to StmtSRefNode |
Ctvm::tir::TensorIntrin | Managed reference to TensorIntrinNode |
Ctvm::tir::Trace | Managed reference to TraceNode |
Ctvm::tir::usmp::AllocatedPoolInfo | |
Ctvm::tir::usmp::BufferInfo | |
Ctvm::tir::usmp::BufferInfoAnalysis | |
Ctvm::tir::usmp::PoolAllocation | |
►Ctvm::transform::Pass | |
Ctvm::transform::Sequential | |
Ctvm::transform::PassContext | PassContext that is used to configure the pass behavior |
Ctvm::transform::PassInfo | Managed reference class for PassInfoNode |
Ctvm::runtime::ObjectTypeChecker< T > | Type traits for runtime type check during FFI conversion |
Ctvm::runtime::ObjectTypeChecker< Array< T > > | |
Ctvm::runtime::ObjectTypeChecker< Map< K, V > > | |
Ctvm::runtime::ObjectTypeChecker< Variant< FirstVariant, RemainingVariants... > > | |
Ctvm::runtime::ObjectTypeChecker< Variant< OnlyVariant > > | |
Ctvm::OpRegEntry | Helper structure to register operators |
Ctvm::runtime::PackedFuncValueConverter< TObjectRef > | Type trait to specify special value conversion rules from TVMArgValue and TVMRetValue |
Ctvm::runtime::PackedFuncValueConverter< Array< T > > | |
Ctvm::runtime::PackedFuncValueConverter< Map< T, U > > | |
Ctvm::runtime::PackedFuncValueConverter< Optional< T > > | |
Ctvm::runtime::PackedFuncValueConverter< PrimExpr > | |
Ctvm::runtime::PackedFuncValueConverter< runtime::Int > | |
Ctvm::runtime::PackedFuncValueConverter< tvm::Bool > | |
Ctvm::runtime::PackedFuncValueConverter< tvm::FloatImm > | |
Ctvm::runtime::PackedFuncValueConverter< tvm::Integer > | |
Ctvm::runtime::PackedFuncValueConverter< tvm::IntImm > | |
Ctvm::runtime::PackedFuncValueConverter< tvm::tir::StringImm > | |
Ctvm::runtime::PackedFuncValueConverter< Variant< VariantTypes... > > | |
Ctvm::runtime::PackedFuncValueConverter<::tvm::runtime::String > | |
Ctvm::relax::PairCons | Constraint of a DFPattern edge (producer -> consumer) in graph-level matching |
Ctvm::runtime::relax_vm::NDArrayCacheMetadata::FileRecord::ParamRecord | Metadata of each parameter |
Ctvm::relay::PatternFunctor< FType > | 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 |
►Ctvm::relay::PatternFunctor< Pattern(const Pattern &)> | |
Ctvm::relay::PatternMutator | A wrapper around ExprFunctor which functionally updates the AST |
Ctvm::relay::PatternFunctor< R(const Pattern &n, Args...)> | |
►Ctvm::relay::PatternFunctor< void(const Pattern &n)> | |
Ctvm::relay::PatternVisitor | A simple visitor wrapper around PatternFunctor |
Ctvm::runtime::profiling::Profiler | |
Ctvm::ReflectionVTable | Virtual function table to support IR/AST node reflection |
Ctvm::ReflectionVTable::Registry | Registry of a reflection table |
Ctvm::runtime::Registry | Registry for global function |
Ctvm::ReprLegacyPrinter | Legacy behavior of ReprPrinter |
►Ctvm::ReprPrinter | A printer class to print the AST/IR nodes |
Ctvm::relay::DFPatternPrinter | A printer class to print pattern |
Ctvm::runtime::ReverseIterAdapter< Converter, TIter > | Iterator adapter that adapts TIter to return another type |
Ctvm::arith::RewriteSimplifier | Rewrite-rule based simplifier |
Ctvm::relay::RuntimeRegEntry | Helper structure to register Runtimes |
Ctvm::te::ScheduleContext | Context helper to collect debug information of Schedule |
Ctvm::meta_schedule::ScopedTimer | |
Ctvm::auto_scheduler::SearchPolicyKey | Attribute keys of ops used for SearchPolicy |
Ctvm::detail::SelectSEqualReduce< T, TraitName, bool > | |
Ctvm::detail::SelectSEqualReduce< T, TraitName, false > | |
Ctvm::detail::SelectSHashReduce< T, TraitName, bool > | |
Ctvm::detail::SelectSHashReduce< T, TraitName, false > | |
Ctvm::detail::SelectVisitAttrs< T, TraitName, bool > | |
Ctvm::detail::SelectVisitAttrs< T, TraitName, false > | |
Ctvm::SEqualReducer | A Reducer class to reduce the structural equality result of two objects |
Ctvm::SHashReducer | A Reducer class to reduce the structural hash value |
Ctvm::SIBuilder | Source Information Builder, SIBuilder provides helper APIs for filling spans, particularly useful for one-to-many, many-to-one and many-to-many IR transformations |
Ctvm::runtime::SignaturePrinter< TSignature > | |
Ctvm::te::Tensor::Slice | 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 |
Ctvm::support::Span< T, W > | A partial implementation of the C++20 std::span |
Ctvm::auto_scheduler::StageAttributes | Stage-level attributes |
Ctvm::tir::StmtFunctor< FType > | Same as ExprFunctor except it is applied on statements |
Ctvm::tir::StmtFunctor< R(const Stmt &n, Args... args)> | |
►Ctvm::tir::StmtFunctor< Stmt(const Stmt &)> | |
►Ctvm::tir::StmtMutator | StmtMutator that mutates the statements |
Ctvm::tir::StmtExprMutator | Mutator that recursively mutates stmts and exprs on them |
►Ctvm::tir::StmtFunctor< void(const Stmt &)> | |
►Ctvm::tir::StmtVisitor | StmtVisitor |
Ctvm::tir::SRefTreeCreator | |
Ctvm::tir::StmtExprVisitor | Visitor that recursively visit stmts and exprs on them |
Ctvm::relax::StructInfoFunctor< FStructInfo > | |
Ctvm::relax::StructInfoFunctor< R(const StructInfo &n, Args...)> | |
►Ctvm::relax::StructInfoFunctor< StructInfo(const StructInfo &n)> | |
Ctvm::relax::StructInfoMutator | StructInfoMutator that mutates struct info |
►Ctvm::relax::StructInfoFunctor< void(const StructInfo &n)> | |
Ctvm::relax::StructInfoVisitor | A struct info visitor |
Ctvm::TargetKindRegEntry | Helper structure to register TargetKind |
Ctvm::TargetTagRegEntry | |
Ctvm::te::TensorDom | Temporary data structure to store union of bounds of each axis of Tensor |
Ctvm::runtime::threading::ThreadGroup | A platform-agnostic abstraction for managing a collection of thread pool threads |
Ctvm::runtime::ThreadLocalDiscoWorker | A threadlocal wrapper of DiscoWorker |
Ctvm::arith::TransitiveComparisonAnalyzer | Using previously specified knowns, compare the expressions provided |
►Cstd::true_type | |
Ctvm::detail::is_specialized< Container< Args... >, Container > | |
Ctvm::runtime::TVMArgs | Arguments into TVM functions |
Ctvm::runtime::TVMArgsSetter | |
Ctvm::runtime::TVMArgsSetterApply< i, T > | |
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 |
CTVMMetadata | Top-level metadata structure. Holds all other metadata types |
Ctvm::runtime::TVMMovableArgValueWithContext_ | Internal auxiliary struct for TypedPackedFunc to indicate a movable argument with additional context information (function name and argument index) for better error reporting |
CTVMParallelGroupEnv | Environment for TVM parallel task |
►Ctvm::runtime::TVMPODValue_ | Internal base class to handle conversion to POD values |
►Ctvm::runtime::TVMPODValue_CRTP_< TVMMovableArgValue_ > | |
Ctvm::runtime::TVMMovableArgValue_ | Internal auxiliary struct for TypedPackedFunc to indicate a movable argument |
►Ctvm::runtime::TVMPODValue_CRTP_< TVMRetValue > | |
Ctvm::runtime::TVMRetValue | Return Value container, Unlike TVMArgValue, which only holds reference and do not delete the underlying container during destruction |
►Ctvm::runtime::TVMPODValue_CRTP_< TVMArgValue > | |
Ctvm::runtime::TVMArgValue | A single argument value to PackedFunc. Containing both type_code and TVMValue |
Ctvm::runtime::TVMPODValue_CRTP_< Derived > | A utility class that adds methods useful for each POD type |
Ctvm::TVMScriptPrinter | Legacy behavior of ReprPrinter |
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 |
Ctvm::TypedEnvFunc< FType > | Please refer to TypedEnvFunc<R(Args..)> |
Ctvm::TypedEnvFunc< bool(const Array< Type > &args, int num_inputs, const Attrs &attrs, const TypeReporter &reporter)> | |
Ctvm::runtime::TypedPackedFunc< FType > | Please refer to TypedPackedFunc<R(Args..)> |
Ctvm::runtime::TypedPackedFunc< Array< ObjectRef >(ObjectRef json_attrs)> | |
Ctvm::runtime::TypedPackedFunc< Array< ObjectRef >(Schedule sch, const Array< ObjectRef > &inputs, const Array< ObjectRef > &attrs, const Optional< ObjectRef > &decision)> | |
Ctvm::runtime::TypedPackedFunc< Array< RunnerResult >(int)> | |
Ctvm::runtime::TypedPackedFunc< Array< te::Tensor >(const Attrs &attrs, const Array< te::Tensor > &inputs, const Type &out_type)> | |
Ctvm::runtime::TypedPackedFunc< Array< TuningRecord >()> | |
Ctvm::runtime::TypedPackedFunc< Array< TuningRecord >(const Workload &, int)> | |
Ctvm::runtime::TypedPackedFunc< Array< tvm::runtime::NDArray >(const TuneContext &context, const Array< MeasureCandidate > &candidates)> | |
Ctvm::runtime::TypedPackedFunc< bool()> | |
Ctvm::runtime::TypedPackedFunc< bool(const IRModule &)> | |
Ctvm::runtime::TypedPackedFunc< int()> | |
Ctvm::runtime::TypedPackedFunc< int64_t()> | |
Ctvm::runtime::TypedPackedFunc< ObjectRef(Array< ObjectRef > attrs)> | |
Ctvm::runtime::TypedPackedFunc< Optional< IRModule >(const IRModule &, const Target &, const String &)> | |
Ctvm::runtime::TypedPackedFunc< Optional< tir::Schedule >(const IRModule &, const Target &, const String &)> | |
Ctvm::runtime::TypedPackedFunc< Optional< TuningRecord >(const IRModule &, const Target &, const String &)> | |
Ctvm::runtime::TypedPackedFunc< R(Args...)> | A PackedFunc wrapper to provide typed function signature. It is backed by a PackedFunc internally |
Ctvm::runtime::TypedPackedFunc< RunnerResult()> | |
Ctvm::runtime::TypedPackedFunc< String()> | |
Ctvm::runtime::TypedPackedFunc< String(const Array< ObjectRef > &inputs, const Array< ObjectRef > &attrs, const Optional< ObjectRef > &decision, const Array< String > &outputs)> | |
Ctvm::runtime::TypedPackedFunc< TargetJSON(TargetJSON)> | |
Ctvm::runtime::TypedPackedFunc< te::Schedule(const Attrs &attrs, const Array< te::Tensor > &outs, const Target &target)> | |
Ctvm::runtime::TypedPackedFunc< tvm::tir::Stmt(Array< tvm::tir::Var > loop_vars, Array< Range > loop_extents, tvm::tir::Stmt loop_body)> | |
Ctvm::runtime::TypedPackedFunc< void()> | |
Ctvm::runtime::TypedPackedFunc< void(Array< TuneContext > tasks, Array< FloatImm > task_weights, int max_trials_global, int max_trials_per_task, int num_trials_per_iter, Builder builder, Runner runner, Array< MeasureCallback > measure_callbacks, Optional< Database > database, Optional< CostModel > cost_model)> | |
Ctvm::runtime::TypedPackedFunc< void(const TaskScheduler &task_scheduler, int task_id, const Array< MeasureCandidate > &measure_candidates, const Array< BuilderResult > &builds, const Array< RunnerResult > &results)> | |
Ctvm::runtime::TypedPackedFunc< void(const TuneContext &, const Array< MeasureCandidate > &, const Array< RunnerResult > &)> | |
Ctvm::runtime::TypedPackedFunc< void(const TuneContext &, const Array< MeasureCandidate > &, void *p_addr)> | |
Ctvm::runtime::TypedPackedFunc< void(const TuningRecord &)> | |
Ctvm::runtime::TypedPackedFunc< void(size_t, void *)> | |
Ctvm::runtime::TypedPackedFunc< void(String)> | |
Ctvm::runtime::TypedPackedFunc< void(tvm::DiagnosticContext ctx)> | |
Ctvm::runtime::TypedPackedFunc< Workload(const IRModule &)> | |
Ctvm::TypeFunctor< FType > | |
Ctvm::TypeFunctor< R(const Type &n, Args...)> | |
►Ctvm::TypeFunctor< Type(const Type &n)> | |
Ctvm::TypeMutator | TypeMutator that mutates expressions |
►Ctvm::TypeFunctor< void(const Type &n)> | |
Ctvm::TypeVisitor | A type visitor that recursively visit types |
Ctvm::runtime::TypeIndex | Namespace for the list of type index |
Ctvm::detail::TypeName< T > | Helper struct to get the type name known to tvm |
Ctvm::detail::TypeName< bool > | |
Ctvm::detail::TypeName< DataType > | |
Ctvm::detail::TypeName< double > | |
Ctvm::detail::TypeName< int > | |
Ctvm::detail::TypeName< int64_t > | |
Ctvm::detail::TypeName< uint64_t > | |
Ctvm::detail::TypeName< void * > | |
Ctvm::relay::v_info | A struct to keep info of traversed expr in ExpandDataflow function |
Ctvm::runtime::Array< T, typename >::ValueConverter | |
Ctvm::detail::ValueTypeInfoMaker< ValueType, IsArray, IsMap > | |
Ctvm::script::printer::IRDocsifierNode::VariableInfo | Information about a variable, including its optional name and its doc creator |
Ctvm::relax::VarUsageInfo | A utility struct returned by CollectVarUsage |
Ctvm::VirtualDeviceCache | A cache of VirtualDevices . This can be used: |
Ctvm::runtime::vm::VMFrame | A representation of a stack frame |
Ctvm::runtime::relax_vm::VMFuncInfo | Information entry in executable function table |
Ctvm::runtime::vm::VMFunction | A representation of a Relay function in the VM |
Ctvm::With< ContextType > | RAII wrapper function to enter and exit a context object similar to python's with syntax |
Ctvm::runtime::WorkerZeroData | A special communication channel between controler and worker-0, assuming they are always collocated in the same process |
Ctvm::meta_schedule::WorkloadEqual | The equality check for Workload |
Ctvm::relax::WorkloadEqual | The equality check for Workload |
Ctvm::meta_schedule::WorkloadHash | The hash method for Workload |
Ctvm::runtime::WrappedPythonObject | A class that wraps a Python object and preserves its ownership |