torch_cuda.RecordContext |
|
torch_cuda.StatType |
|
torch._RegisterOrVerify |
The NoInferSchemaTag is a type name used to indicate that this call to the
CppFunction constructor should not trigger schema inference from functor.
|
torch.ActiveProfilerType |
|
torch.ActivityType |
|
torch.AliasAnalysisKind |
|
torch.AttributeKind |
|
torch.AutogradFallbackMode |
|
torch.Backend |
This legacy enum class defines the set of backends supported by old school,
code generated Type-based ATen.
|
torch.BackendComponent |
|
torch.CreationMeta |
NOTE [ Autograd View Variables ]
Many operations return Variable that shares storage with an input Variable.
|
torch.DebugInfoKind |
|
torch.DeviceType |
|
torch.DispatchKey |
|
torch.EventFlag |
Flags defining the behavior of events.
|
torch.ExecutorExecutionMode |
|
torch.FastSetupType |
|
torch.Float32MatmulPrecision |
|
torch.FusionBehavior |
|
torch.IListRefTag |
|
torch.JitAttributeKind |
|
torch.Layout |
|
torch.LinalgBackend |
|
torch.MemoryFormat |
Construct an ArrayRef from a C array.
|
torch.MemOverlap |
|
torch.MemOverlapStatus |
|
torch.NameType |
|
torch.padding_mode |
|
torch.PickleOpCode |
|
torch.ProfilerState |
|
torch.PyInterpreterStatus |
|
torch.QEngine |
QEngine is an enum that is used to select the engine to run quantized ops.
|
torch.QScheme |
QScheme is an enum that specifies the type of quantization.
|
torch.RecordScope |
|
torch.Reduction |
|
torch.RegistryPriority |
Simple registry implementation that uses static variables to
register object creators during program initialization time.
|
torch.ScalarType |
|
torch.SchemaArgType |
|
torch.Tag |
Include this file if you want to register operators.
|
torch.TensorDataContainerType |
|
torch.TensorIndexType |
|
torch.TokenKind |
|
torch.TorchFunctionDisabledState |
|
torch.TypeKind |
|
torch.TypeVerbosity |
|