extension transpose : IForwardDifferentiable<transpose<T, N, M>>
int8_t4_packed
uint8_t4_packed
Attributes
AutoPyBindCUDA
BackwardDerivative
BackwardDerivativeOf
BackwardDifferentiable
COM
CUDADeviceExport
CUDAHost
CUDAKernel
CudaDeviceExport
CudaHost
CudaKernel
DerivativeGroupLinear
DerivativeGroupQuad
DerivativeMember
Differentiable
DllExport
DllImport
ExperimentalModule
Flags
ForceInline
ForceUnroll
ForwardDerivative
ForwardDerivativeOf
ForwardDifferentiable
HasTrivialForwardDerivative
KnownBuiltin
MaxIters
MaximallyReconverges
MaybeDifferentiable
NoDiffThis
NonUniformReturn
NumThreads
OverloadRank
PreferCheckpoint
PreferRecompute
PrimalSubstitute
PrimalSubstituteOf
PyExport
QuadDerivatives
RemovedSince
RequireFullQuads
RequirePrelude
Shader
SpecializationConstant
Specialize
TorchEntryPoint
TreatAsDifferentiable
UnscopedEnum
WaveSize
allow
allow_uav_condition
anyValueSize
branch
builtin
call
constref
deprecated
disable_array_flattening
domain
earlydepthstencil
fastopt
flatten
forcecase
format
gl_binding
instance
loop
maxtessfactor
maxvertexcount
mutating
noRefInline
noinline
nonmutating
numthreads
open
outputcontrolpoints
outputtopology
partitioning
patchconstantfunc
push_constant
raypayload
require
sealed
shader
shader_record
spv_target_env_1_3
unroll
vk_binding
vk_constant_id
vk_image_format
vk_location
vk_offset
vk_push_constant
vk_shader_record
vk_specialization_constant
vk_spirv_instruction
Global Declarations
Atomic functions
InterlockedAdd
InterlockedAnd
InterlockedCompareExchange
InterlockedCompareExchangeFloatBitwise
InterlockedCompareStore
InterlockedCompareStoreFloatBitwise
InterlockedExchange
InterlockedMax
InterlockedMin
InterlockedOr
InterlockedXor
Memory and control barriers
AllMemoryBarrier
AllMemoryBarrierWithGroupSync
DeviceMemoryBarrier
DeviceMemoryBarrierWithGroupSync
GroupMemoryBarrier
GroupMemoryBarrierWithGroupSync
Bit operation functions
countbits
firstbithigh
firstbitlow
reversebits
Conversion functions
asdouble
asfloat
asfloat16
asint
asint16
asuint
asuint16
bit_cast
f16tof32
f32tof16
f32tof16_
reinterpret
Derivative functions
ddx
ddx_coarse
ddx_fine
ddy
ddy_coarse
ddy_fine
fwidth
fwidth_coarse
fwidth_fine
Vertex Interpolation Functions
EvaluateAttributeAtCentroid
EvaluateAttributeAtSample
EvaluateAttributeSnapped
Math functions
abs
acos
acosh
asin
asinh
atan
atan2
atanh
ceil
clamp
copysign
copysign_double
copysign_float
copysign_half
cos
cosh
cospi
cross
degrees
determinant
distance
divide
dot
dot2add
dot4add_i8packed
dot4add_u8packed
dst
exp
exp10
exp2
fabs
faceforward
fdim
floor
fma
fmax
fmax3
fmedian3
fmin
fmin3
fmod
frac
fract
frexp
isfinite
isinf
isnan
ldexp
length
lerp
lit
log
log10
log2
mad
max
max3
median3
min
min3
modf
msad4
mul
normalize
pow
powr
radians
rcp
reflect
refract
rint
round
rsqrt
saturate
sign
sin
sincos
sinh
sinpi
smoothstep
sqrt
step
tan
tanh
tanpi
transpose
trunc
Mesh shading
DispatchMesh
SetMeshOutputCounts
Ray-tracing
AcceptHitAndEndSearch
CANDIDATE_NON_OPAQUE_TRIANGLE
CANDIDATE_PROCEDURAL_PRIMITIVE
COMMITTED_NOTHING
COMMITTED_PROCEDURAL_PRIMITIVE_HIT
COMMITTED_TRIANGLE_HIT
CallShader
DispatchRaysDimensions
DispatchRaysIndex
GeometryIndex
GetClusterID
GetInstanceTraversable
GetLssPositionsAndRadii
GetSpherePositionAndRadius
GetTransformListHandle
GetTransformListSize
GetTraversableChild
GetTraversableInstanceId
GetTraversableObjectToWorld
GetTraversableTransformType
GetTraversableWorldToObject
HIT_KIND_TRIANGLE_BACK_FACE
HIT_KIND_TRIANGLE_FRONT_FACE
HitKind
HitTriangleVertexPosition
IgnoreHit
InstanceID
InstanceIndex
IsLssHit
IsSphereHit
ObjectRayDirection
ObjectRayOrigin
ObjectToWorld
ObjectToWorld3x4
ObjectToWorld4x3
PrimitiveIndex
RAYQUERY_FLAG_ALLOW_OPACITY_MICROMAPS
RAYQUERY_FLAG_NONE
RAY_FLAG_ACCEPT_FIRST_HIT_AND_END_SEARCH
RAY_FLAG_CULL_BACK_FACING_TRIANGLES
RAY_FLAG_CULL_FRONT_FACING_TRIANGLES
RAY_FLAG_CULL_NON_OPAQUE
RAY_FLAG_CULL_OPAQUE
RAY_FLAG_FORCE_NON_OPAQUE
RAY_FLAG_FORCE_OMM_2_STATE
RAY_FLAG_FORCE_OPAQUE
RAY_FLAG_NONE
RAY_FLAG_SKIP_CLOSEST_HIT_SHADER
RAY_FLAG_SKIP_PROCEDURAL_PRIMITIVES
RAY_FLAG_SKIP_TRIANGLES
RayCurrentTime
RayFlags
RayTCurrent
RayTMin
ReportHit
ReportHitOptix
TraceMotionRay
TraceRay
WorldRayDirection
WorldRayOrigin
WorldToObject
WorldToObject3x4
WorldToObject4x3
Tessellation functions
Process2DQuadTessFactorsAvg
Process2DQuadTessFactorsMax
Process2DQuadTessFactorsMin
ProcessIsolineTessFactors
ProcessQuadTessFactorsAvg
ProcessQuadTessFactorsMax
ProcessQuadTessFactorsMin
ProcessTriTessFactorsAvg
ProcessTriTessFactorsMax
ProcessTriTessFactorsMin
Wave and quad functions
QuadReadAcrossDiagonal
QuadReadAcrossX
QuadReadAcrossY
QuadReadLaneAt
WaveActiveAllEqual
WaveActiveAllTrue
WaveActiveAnyTrue
WaveActiveBallot
WaveActiveBitAnd
WaveActiveBitOr
WaveActiveBitXor
WaveActiveCountBits
WaveActiveMax
WaveActiveMin
WaveActiveProduct
WaveActiveSum
WaveBroadcastLaneAt
WaveGetActiveMulti
WaveGetConvergedMulti
WaveGetLaneCount
WaveGetLaneEqMask
WaveGetLaneGeMask
WaveGetLaneGtMask
WaveGetLaneIndex
WaveGetLaneLeMask
WaveGetLaneLtMask
WaveGetNumWaves
WaveIsFirstLane
WaveMatch
WaveMultiPrefixCountBits
WavePrefixBitAnd
WavePrefixBitOr
WavePrefixBitXor
WavePrefixCountBits
WavePrefixMax
WavePrefixMin
WavePrefixProduct
WavePrefixSum
WaveReadLaneAt
WaveReadLaneFirst
WaveShuffle
_WaveCountBits
CheckAccessFullyMapped
D3DCOLORtoUBYTE4
GetAttributeAtVertex
GetRenderTargetSampleCount
GetRenderTargetSamplePosition
InterlockedAddF16Emulated
InterlockedAddF16x2
IsHelperLane
MaybeReorderThread
NonUniformResourceIndex
QuadAll
QuadAny
ReorderThread
WaveClusteredRotate
WaveMultiBitAnd
WaveMultiBitOr
WaveMultiBitXor
WaveMultiMax
WaveMultiMin
WaveMultiPrefixBitAnd
WaveMultiPrefixBitOr
WaveMultiPrefixBitXor
WaveMultiPrefixExclusiveBitAnd
WaveMultiPrefixExclusiveBitOr
WaveMultiPrefixExclusiveBitXor
WaveMultiPrefixExclusiveMax
WaveMultiPrefixExclusiveMin
WaveMultiPrefixExclusiveProduct
WaveMultiPrefixExclusiveSum
WaveMultiPrefixInclusiveBitAnd
WaveMultiPrefixInclusiveBitOr
WaveMultiPrefixInclusiveBitXor
WaveMultiPrefixInclusiveMax
WaveMultiPrefixInclusiveMin
WaveMultiPrefixInclusiveProduct
WaveMultiPrefixInclusiveSum
WaveMultiPrefixProduct
WaveMultiPrefixSum
WaveMultiProduct
WaveMultiSum
WaveRotate
WorkgroupCount
WorkgroupSize
abort
all
any
bitfieldExtract
bitfieldInsert
clip
clock2x32ARB
clockARB
concat
coopVecLoad
coopVecLoadCoherent
coopVecLoadGroupshared
coopVecMatMul
coopVecMatMulAdd
coopVecMatMulAddPacked
coopVecMatMulPacked
coopVecOuterProductAccumulate
coopVecReduceSumAccumulate
createDynamicObject
cudaBlockDim
cudaBlockIdx
cudaThreadIdx
debugBreak
defaultGetDescriptorFromHandle
detach
diffPair
enableVMMDeviceScopeCapabilityIfNeeded
floatCast
getDescriptorFromHandle
getRealtimeClock
getRealtimeClockLow
getStringHash
loadAligned
makeArrayFromElement
makeTuple
nextafter
nonuniform
operator*
operator?:
packHalf2x16
packInt4x8
packInt4x8Clamp
packSnorm2x16
packSnorm4x8
packUint4x8
packUint4x8Clamp
packUnorm2x16
packUnorm4x8
pack_clamp_s8
pack_clamp_u8
pack_s8
pack_u8
printf
select
static_assert
storeAligned
syncTorchCudaStream
unmodified
unpackHalf2x16ToFloat
unpackHalf2x16ToHalf
unpackInt4x8ToInt16
unpackInt4x8ToInt32
unpackSnorm2x16ToFloat
unpackSnorm2x16ToHalf
unpackSnorm4x8ToFloat
unpackSnorm4x8ToHalf
unpackUint4x8ToUint16
unpackUint4x8ToUint32
unpackUnorm2x16ToFloat
unpackUnorm2x16ToHalf
unpackUnorm4x8ToFloat
unpackUnorm4x8ToHalf
unpack_s8s16
unpack_s8s32
unpack_u8u16
unpack_u8u32
unused
updateDiff
updatePair
updatePrimal
workgroupUniformLoad
interface IAtomicable
Description
Represents types that can be used in any atomic operations.
Implemented by builtin scalar types: int, uint, int64_t, uint64_t, int8_t, uint8_t, int16_t, uint16_t, float, double and half.