DynamicSlice

#include <popops/DynamicSlice.hpp>

Support for dynamic slices.

namespace poplar: Poplar classes and functions.

namespace popops

Common functions, such as elementwise and reductions.

map

Map an expression across tensors.

Elementwise Options

enableGenerateCodelet (true, false) [=true]

When true and the following conditions are met, poplar will generate a codelet to execute the map operation. Otherwise, it will sequence poplibs codelets to create the expression.
- All of the inputs are of the same size
- Inputs do not alias
- Multiple operations are being performed

mapInPlace

Update the input tensors with the result of map().

mapWithOutput

Write the result of map() to the given output tensor.

checkTypes

Check that the host compile-time type constType is compatible with the run-time IPU type elementType.

param elementType: The run-time IPU type.
param constant: Unused.

tparam constType: The host compile-time type.

throws std::runtime_error: If the types are not compatible.

varianceToInvStdDev

Convert variance to inverse standard deviation.

Each element in the output tensor is the result of 1 / sqrt(variance_value + epsilon), where variance_value is the corresponding element in variance.

Warning

If variance_value + epsilon is zero then the result will be invalid and this operation could generate a divide-by-zero floating-point exception (if enabled).

param graph: The graph to update.
param variance: A tensor of variance values.
param epsilon: A (typically small) scalar to add to the variance values, to avoid numerical issues (for example, divide by zero).
param prog: The sequence of programs to append this conversion operation to.
param debugContext: Optional debug information.

return: A tensor where each element is the inverse standard deviation.

invStdDevToVariance

Convert inverse standard deviation to variance.

Each element in the output tensor is the result of 1 / (invStdDev_value + epsilon)^2, where invStdDev_value is the corresponding element in invStdDev.

Warning

If invStdDev_value + epsilon is zero then the result will be invalid and this operation could generate a divide-by-zero floating-point exception (if enabled).

param graph: The graph to update.
param invStdDev: A tensor of inverse standard deviation values.
param epsilon: A (typically small) scalar to add to the variance values, to avoid numerical issues (for example, divide by zero).
param prog: The sequence of programs to append this conversion operation to.
param debugContext: Optional debug information.
param options: A list of flags to pass to the expression evaluator.

return: A tensor where each element is the variance.

add

Add each element in A to the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a + b, where a and b are the corresponding elements of A and B tensors respectively.

addInPlace

Update the tensor A with the result of add().

See add() for parameter descriptions.

addWithOutput

Write the result of add() to the given output tensor, out.

See add() for the remaining parameter descriptions.

param out: The tensor to write the results to.

sub

Subtract the elements of B from A and return the result in a new tensor.

param graph: The graph to update.
param A: The tensor of elements which will be subtracted from.
param B: The tensor of elements to subtract from A.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is equal to a - b, where a and b are the corresponding elements of A and B tensors respectively.

subInPlace

Update the tensor A with the result of sub().

See sub() for parameter descriptions.

subWithOutput

Write the result of sub() to the given output tensor, out.

See sub() for the remaining parameter descriptions.

param out: The tensor to write the results to.

mul

Multiply each element in A by the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of a * b, where a and b are the corresponding elements of A and B tensors respectively.

mulInPlace

Update the tensor A with the result of mul().

See mul() for parameter descriptions.

mulWithOutput

Write the result of mul() to the given output tensor, out.

See mul() for the remaining parameter descriptions.

param out: The tensor to write the results to.

div

Divide each element in A by the corresponding element in B.

param graph: The graph to update.
param A: The tensor of dividends.
param B: The tensor of divisors.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a / b, where a and b are the corresponding elements of A and B tensors respectively.

divInPlace

Update the tensor A with the result of div().

See div() for parameter descriptions.

divWithOutput

Write the result of div() to the given output tensor, out.

See div() for the remaining parameter descriptions.

param out: The tensor to write the results to.

pow

Compute each element in A to the power of the corresponding element in B.

param graph: The graph to update.
param A: The tensor of bases.
param B: The tensor of exponents.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is equal to pow(a, b), where a and b are the corresponding elements of A and B tensors respectively.

powInPlace

Update the tensor A with the result of pow().

See pow() for parameter descriptions.

powWithOutput

Write the result of pow() to the given output tensor, out.

See pow() for the remaining parameter descriptions.

param out: The tensor to write the results to.

rem

Compute the remainder of each element in A divided by the corresponding element in B.

param graph: The graph to update.
param A: The tensor of dividends.
param B: The tensor of divisors.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is equal to a % b, where a and b are the corresponding elements of A and B tensors respectively.

remInPlace

Update the tensor A with the result of rem().

See rem() for parameter descriptions.

remWithOutput

Write the result of rem() to the given output tensor, out.

See rem() for the remaining parameter descriptions.

param out: The tensor to write the results to.

bitwiseAnd

Compute the bitwise AND of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a & b, where a and bare the corresponding elements of A and B tensors respectively.

bitwiseAndInPlace

Update the tensor A with the result of bitwiseAnd().

See bitwiseAnd() for parameter descriptions.

bitwiseAndWithOutput

Write the result of bitwiseAnd() to the given output tensor, out.

See bitwiseAnd() for the remaining parameter descriptions.

param out: The tensor to write the results to.

bitwiseOr

Compute the bitwise OR of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a | b, where a and b are the corresponding elements of A and B tensors respectively.

bitwiseOrInPlace

Update the tensor A with the result of bitwiseOr().

See bitwiseOr() for parameter descriptions.

bitwiseOrWithOutput

Write the result of bitwiseOr() to the given output tensor, out.

See bitwiseOr() for the remaining parameter descriptions.

param out: The tensor to write the results to.

bitwiseXor

Compute the bitwise XOR of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a ^ b, where a and b are the corresponding elements of A and B tensors respectively.

bitwiseXorInPlace

Update the tensor A with the result of bitwiseXor().

See bitwiseXnor() for parameter descriptions.

bitwiseXorWithOutput

Write the result of bitwiseXor() to the given output tensor, out.

See bitwiseXor() for the remaining parameter descriptions.

param out: The tensor to write the results to.

bitwiseXnor

Compute the bitwise XNOR of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of !(a ^ b), where a and b are the corresponding elements of A and B tensors respectively.

bitwiseXnorInPlace

Update the tensor A with the result of bitwiseXnor().

See bitwiseXnor() for parameter descriptions.

bitwiseXnorWithOutput

Write the result of bitwiseXnor() to the given output tensor, out.

See bitwiseXnor() for the remaining parameter descriptions.

param out: The tensor to write the results to.

shiftLeft

Shift the elements of A left by the corresponding elements of B.

param graph: The graph to update.
param A: The tensor of elements which to left-shift.
param B: The tensor of elements that describe the amount to left-shift A by.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is equal to a << b, where a and b are the corresponding elements of A and B tensors respectively.

shiftLeftInPlace

Update the tensor A with the result of shiftLeft().

See shiftLeft() for parameter descriptions.

shiftLeftWithOutput

Write the result of shiftLeft() to the given output tensor, out.

See shiftLeft() for the remaining parameter descriptions.

param out: The tensor to write the results to.

shiftRight

Shift the elements of A right by the corresponding elements of B.

param graph: The graph to update.
param A: The tensor of elements which to right-shift.
param B: The tensor of elements that describe the amount to right-shift by. A.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is equal to a >> b (without sign extension), where a and b are the corresponding elements of A and B tensors respectively.

shiftRightInPlace

Update the tensor A with the result of shiftRight().

See shiftRight() for parameter descriptions.

shiftRightWithOutput

Write the result of shiftRight() to the given output tensor, out.

See shiftRight() for the remaining parameter descriptions.

param out: The tensor to write the results to.

shiftRightSignExtend

Shift the elements of A right with sign extension by the corresponding elements of B.

param graph: The graph to update.
param A: The tensor of elements which to right-shift.
param B: The tensor of elements that describe the amount to right-shift A by.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is equal to a >> b with sign extension, where a and b are the corresponding elements of A and B tensors respectively.

shiftRightSignExtendInPlace

Update the tensor A with the result of shiftRightSignExtend().

See shiftRightSignExtend() for parameter descriptions.

shiftRightSignExtendWithOutput

Write the result of shiftRightSignExtend() to the given output tensor, out.

See shiftRightSignExtend() for the remaining parameter descriptions.

param out: The tensor to write the results to.

logicalAnd

Compute the logical AND (&&) of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of a && b, where a and b are the corresponding elements of A and B tensors respectively.

logicalAndInPlace

Update the tensor A with the result of logicalAnd().

See logicalAnd() for parameter descriptions.

logicalAndWithOutput

Write the result of logicalAnd() to the given output tensor, out.

See logicalAnd() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

logicalOr

Compute the logical OR (||) of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of a || b, where a and b are the corresponding elements of A and B tensors respectively.

logicalOrInPlace

Update the tensor A with the result of logicalOr().

See logicalOr() for parameter descriptions.

logicalOrWithOutput

Write the result of logicalOr() to the given output tensor, out.

See logicalOr() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

eq

Check if each element in A is equal to the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a == b, where a and b are the corresponding elements of A and B tensors respectively.

eqInPlace

Update the tensor A with the result of eq().

See eq() for parameter descriptions.

eqWithOutput

Write the result of eq() to the given output tensor, out.

See eq() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

neq

Check if each element in A is not equal to the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of a != b, where a and b are the corresponding elements of A and B tensors respectively.

neqInPlace

Update the tensor A with the result of neq().

See neq() for parameter descriptions.

neqWithOutput

Write the result of neq() to the given output tensor, out.

See neq() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

gteq

Check if each element in A is greater than or equal to the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a >= b, where a and b are the corresponding elements of A and B tensors respectively.

gteqInPlace

Update the tensor A with the result of gteq().

See gteq() for parameter descriptions.

gteqWithOutput

Write the result of gteq() to the given output tensor, out.

See gteq() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

gt

Check if each element in A is greater than the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of a > b, where a and b are the corresponding elements of A and B tensors respectively.

gtInPlace

Update the tensor A with the result of gt().

See gt() for parameter descriptions.

gtWithOutput

Write the result of gt() to the given output tensor, out.

See gt() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

lteq

Check if each element in A is less than or equal to the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of a <= b, where a and b are the corresponding elements of A and B tensors respectively.

lteqInPlace

Update the A with the result of lteq().

See lteq() for parameter descriptions.

lteqWithOutput

Write the result of lteq() to the given output tensor, out.

See lteq() for the remaining parameter descriptions.

param out: The tensor to write the booleans to.

lt

Check if each element in A is less than the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of a < b, where a and b are the corresponding elements of A and B tensors respectively.

ltInPlace

Update the A with the result of lt().

See lt() for parameter descriptions.

ltWithOutput

Write the result of lt() to the given output tensor, out.

See lt() for the remaining parameter descriptions.

param out: The tensor to write the boolean results to.

max

Compute the maximum of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of max(a, b), where a and b are the corresponding elements of A and B tensors respectively.

maxInPlace

Update the tensor A with the result of max().

See max() for parameter descriptions.

maxWithOutput

Write the result of max() to the given output tensor, out.

See max() for the remaining parameter descriptions.

param out: The tensor to write the maximums to.

min

Compute the minimum of each element in A with the corresponding element in B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information.
param options: Element-wise options. See map().

return: A tensor where each element is the result of min(a, b), where a and b are the corresponding elements of A and B tensors respectively.

minInPlace

Update the tensor A with the result of min().

See min() for parameter descriptions.

minWithOutput

Write the result of min() to the given output tensor, out.

See min() for the remaining parameter descriptions.

param out: The tensor to write the minimums to.

atan2

Compute the element-wise arctangent of A / B.

param graph: The graph to update.
param A: A tensor of elements.
param B: A tensor of elements.
param prog: The sequence to extend with the execution of the expression evaluation.
param debugContext: Optional debug information
param options: Element-wise options. See map().

return: A tensor where each element is the result of arctan(a / b); a and b are the corresponding elements of A and B tensors respectively.

atan2InPlace

Update the tensor A with the result of atan2().

See atan2() for parameter descriptions.

atan2WithOutput

Write the result of atan2() to the given output tensor, out.

See atan2() for the remaining parameter descriptions.

param out: The tensor to write the result to.

invStdDevToVarianceInPlace

Update the invStdDev tensor with the result of invStdDevToVariance().

See invStdDevToVariance() for parameter descriptions.

invStdDevToVarianceWithOutput

Write the result of invStdDevToVariance() to the given output tensor, out.

See invStdDevToVariance() for the remaining parameter descriptions.

param out: The tensor to write the variance to.

varianceToInvStdDevInPlace

Update the variance tensor with the result of varianceToInvStdDev().

See varianceToInvStdDev() for parameter descriptions.

varianceToInvStdDevWithOutput

Write the result of varianceToInvStdDev() to the given output tensor, out.

See varianceToInvStdDev() for the remaining parameter descriptions.

param out: The tensor to write inverse standard deviation to.

Functions

bool operator<(const SlicePlan &a, const SlicePlan &b) noexcept

bool operator==(const SlicePlan &a, const SlicePlan &b) noexcept

bool operator!=(const SlicePlan &a, const SlicePlan &b) noexcept

poplar::Tensor createSliceableTensor(poplar::Graph &graph, const poplar::Type &type, const std::vector<size_t> &shape, const std::vector<size_t> &dims, const std::vector<size_t> &sizes, std::size_t minGrainSize = 0, const poplar::DebugContext &debugContext = {})

Create and map a tensor to be sliced/updated efficiently.

The returned tensor will be spread over as many tiles as possible while respecting the minimum number of elements per tile (minGrainSize) and still being in a form that can be sliced/updated efficiently.

Parameters

graph – The Poplar graph.
type – The type of the elements.
shape – The shape of the tensor to be slice/updated.
dims – The dimensions of the tensor that will be slice/updated.
sizes – The size of the slice in each of the dimensions.
minGrainSize – The minimum elements per slice mapped to each tile
debugContext – Optional debug information.

Returns

A tensor shape shape that is suitably mapped

poplar::Tensor createSliceableTensor(poplar::Graph &graph, const poplar::Type &type, const std::vector<size_t> &shape, const std::vector<size_t> &dims, const std::vector<size_t> &sizes, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Create and map a tensor to be sliced/updated efficiently.

The returned tensor will be laid out according to the plan.

Parameters

graph – The Poplar graph.
type – The type of the elements.
shape – The shape of the tensor to be slice/updated.
dims – The dimensions of the tensor that will be slice/updated.
sizes – The size of the slice in each of the dimensions.
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor shape shape that is suitably mapped.

poplar::Tensor createGroupedSliceableTensor(poplar::Graph &graph, const poplar::Type &type, const std::size_t groupSize, const std::vector<std::size_t> &shape, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Create and map a tensor with a group dimension to be sliced/updated efficiently.

Groups allow multiple independent slice/update operations of the same size to be done in parallel.

The shape, dims, and sizes parameters are defined as in createSliceableTensor() and are used for every group.

A valid plan must be provided. The returned tensor will be laid out according to the plan.

Parameters

graph – The Poplar graph.
type – The type of the elements.
groupSize – The group size.
shape – The shape of the tensor to be slice/updated.
dims – The dimensions of the tensor that will be slice/updated.
sizes – The size of the slice in each of the dimensions.
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor suitably mapped with groupSize as the first dimension and the remaining dimensions as in shape.

poplar::Tensor createSliceTensor(poplar::Graph &graph, const poplar::Tensor &t, const std::vector<size_t> &dims, const std::vector<size_t> &sizes, std::size_t numIndices, const poplar::DebugContext &debugContext = {})

Create and map a tensor that is used as a result of slicing, or as an input to an update.

Introspection on the tensor t is used to lay out the created tensor such that it can be used to efficiently update t.

Parameters

graph – The Poplar graph.
t – The tensor to be updated.
dims – The dimensions of the tensor that will be sliced/updated.
sizes – The number of elements of each dimension in dims that will be sliced/updated.
numIndices – The number of slices this tensor should contain.
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor with shape [numIndices, shape…] mapped appropriately to be sliced into/updated from.

poplar::Tensor createSliceTensor(poplar::Graph &graph, const poplar::Type &type, const std::vector<std::size_t> &shape, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, std::size_t numIndices, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Create and map a tensor that is used as a result of slicing, or as an input to an update.

The returned tensor is laid out according to the plan for the slice/update operation.

Parameters

graph – The Poplar graph.
type – The type of the elements.
shape – The shape of the tensor to be slice/updated.
dims – The dimensions of the tensor that will be sliced/updated.
sizes – The number of elements of each dimension in dims that will be sliced/updated.
numIndices – The number of slices this tensor should contain.
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor with shape [numIndices, shape…]

poplar::Tensor createGroupedSliceTensor(poplar::Graph &graph, const poplar::Type &type, const std::size_t groupSize, const std::vector<std::size_t> &shape, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, const std::size_t numIndices, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Create and map a tensor with a group dimension that is used as a result of slicing, or as an input to an update.

The group dimension is the first dimension of the output tensor.

Groups allow multiple independent slice/update operations of the same size to be done in parallel.

The shape, dims, and sizes parameters are defined as in createGroupedSliceTensor and are used for every group.

The returned tensor is laid out according to the plan for the slice/update operation. It is an error to call this function without a valid plan.

Parameters

graph – The Poplar graph.
type – The type of the elements.
groupSize – The group size
shape – The shape of the tensor to be slice/updated.
dims – The dimensions of the tensor that will be sliced/updated.
sizes – The number of elements of each dimension in dims that will be sliced/updated.
numIndices – The number of slices this tensor should contain.
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor with shape [groupSize, numIndices, shape…]

poplar::Tensor createIndicesTensor(poplar::Graph &graph, const std::vector<std::size_t> &dims, std::size_t numIndices, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Create and map a tensor to contain indices for slicing or updating a tensor efficiently.

Parameters

graph – The Poplar graph.
dims – The dimensions of a tensor to be sliced/updated that will be sliced/updated using these indices.
numIndices – The number of indices this tensor should contain
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor of shape [numIndices, dims.size()] mapped appropriately to be used as the indices for a slice/update operation. Element type is always UNSIGNED_INT.

poplar::Tensor createGroupedIndicesTensor(poplar::Graph &graph, const std::size_t groupSize, const std::vector<std::size_t> &dims, const std::size_t numIndices, const SlicePlan &plan, const poplar::OptionFlags&, const poplar::DebugContext &debugContext = {})

Create and map a tensor with a group dimension to contain indices for slicing or updating a tensor efficiently.

Groups allow multiple independent slice/update operations of the same size to be done in parallel. Indices have the same bounds for every element in the group.

dims is defined as in the non-grouped version. They do not include a group dimension.

It is an error to call this function without a valid plan and groupSize must match the group size with which the operation is planned in plan.

Parameters

graph – The Poplar graph.
groupSize – The size of the group
dims – The dimensions of a tensor to be sliced/updated that will be sliced/updated using these indices.
numIndices – The number of indices this tensor should contain
plan – Plan describing how the slicing/updating operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

Returns

A tensor of shape [groupSize, numIndices, dims.size()] mapped appropriately to be used as the indices for a slice/update operation. Element type is always UNSIGNED_INT.

poplar::Tensor createSliceableTensorFromSlice(poplar::Graph &graph, const poplar::Tensor &s, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &numSlices, const poplar::DebugContext &debugContext = {})

Create and map a tensor to be sliced/updated.

The tensor is mapped in a way that can be efficiently sliced and updated to/from the given slice tensor. It will be distributed across as many tiles as the given slice and with the same contiguous regions on each tile. The tensor’s shape and mapping are derived from the reference slice tensor.

Parameters

graph – The Poplar graph.
s – The reference slice.
dims – The dimensions of the returned tensor that will be sliced.
numSlices – The number of independent slices in each sliced dimension.
debugContext – Optional debug information.

Returns

A tensor to be sliced/updated.

poplar::Tensor dynamicSlice(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &offset, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Slice a tensor based on offsets specified by a tensor.

dims gives the dimensions to slice, sizes defines the size of the slice in those dimensions and offset gives the base offsets on each execution.

offset[0], dims and sizes must have the same size. offset may have a second dimension with an element per tile, which can eliminate exchange.

** Dynamic slice options **

remapOutOfBoundIndices (true, false) [=false] Out of bounds indices are mapped to index 0.
paddingIndexUsed (true, false) [=false] Padding index equal to the size of the slice dimension of tensor t may be used in the indices. The actual padding values returned for a padding index are zeros.

Note

If the elements of a single slice cannot be well spread across available tiles (for example, because it has fewer elements than the available number of tiles) then the operation as a whole will also have poor tile balance and consequently poor tile memory balance.

Parameters

graph – The Poplar graph.
t – The source tensor.
offset – A tensor of offsets at which the output is extracted.
dims – The dimensions of t to slice.
sizes – The size of the slice in each of the dimensions in dims.
prog – The program to be extended
options – Option flags
debugContext – Optional debug information.

Returns

The specified subtensor

poplar::Tensor createDynamicSliceOutput(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &offset, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Create a tensor for use with dynamicSliceWithOutput.

Parameters, allowed options and the return values are the same as dynamicSlice except there is no prog parameter.

Parameters

graph – The Poplar graph.
t – The source tensor.
offset – A tensor of offsets at which the output is extracted.
dims – The dimensions of t to slice.
sizes – The size of the slice in each of the dimensions in dims.
options – Option flags
debugContext – Optional debug information.

Returns

The specified subtensor

void dynamicSliceWithOutput(poplar::Graph &graph, const poplar::Tensor &output, const poplar::Tensor &t, const poplar::Tensor &offset, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Slice a tensor based on offsets specified by a tensor.

dims gives the dimensions to slice, sizes defines the size of the slice in those dimensions and offset gives the base offsets on each execution.

offset[0], dims and sizes must have the same size. offset may have a second dimension with an element per tile, which can eliminate exchange.

see dynamicSlice for information on options

Parameters

graph – The Poplar graph.
output – The output tensor, This should ideally be created with createSliceTensor to maximise efficiency,
t – The source tensor.
offset – A tensor of offsets at which the output is extracted.
dims – The dimensions of t to slice.
sizes – The size of the slice in each of the dimensions in dims.
prog – The program to be extended
debugContext – Optional debug information.
debugContext – Optional debug information
options – Option to control remapping of indices and padding

poplar::Graph::TileToTensorMapping getSliceMapping(poplar::Graph &graph, const poplar::Tensor &t, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes)

Get the tile mapping for a slice of a tensor.

dims gives the dimensions to slice, sizes defines the size of the slice in those dimensions.

Parameters

graph – The Poplar graph.
t – The source tensor.
dims – The dimensions of t to slice.
sizes – The size of the slice in each of the dimensions in dims.

void dynamicUpdate(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offset, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &options = {})

Update a subtensor at offsets read from a tensor.

dims gives the dimensions that are partially updated, by sizes elements, at offsets offset. Unspecified dimensions are copied in full with zero offset.

offset[0], dims and sizes must have the same size. offset may have a second dimension with an element per tile, which can eliminate exchange.

see dynamicSlice for information on options

Parameters

graph – The Poplar graph.
t – The tensor to update.
s – The updates.
offset – The offset within t to be updated.
dims – The dimensions to be dynamically updated.
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
debugContext – Optional debug information.
options – Option flags

poplar::Tensor multiSlice(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &offsets, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Take multiple slices from a base tensor.

The returned tensor will have a rank one greater than t. Its outer dimension will be offsets.dim(0). Note that dims refers to the dimensions of t. t can be created using createSliceableTensor() to ensure efficient mapping.

dims and sizes must be size 1.

** multiSlice options **

remapOutOfBoundIndices (true, false) [=false] Out of bounds indices are mapped to index 0.
paddingIndexUsed (true, false) [=false] Padding index equal to the size of the slice dimension of tensor t may be used in the indices. The actual padding values returned for a padding index are zeros. Padding index is excluded from index validation if it is enabled.

Parameters

graph – The Poplar graph.
t – The tensor being sliced.
offsets – The offsets within t to be sliced.
dims – The dimensions of t to be sliced.
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

poplar::Tensor groupedMultiSlice(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &offsets, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Take multiple slices from a base tensor where the base tensor and the offsets tensor have the group dimension as the first dimension.

The indices given in the offsets tensor should be in the range [0, base.dim(1)]. Indices outside of this range are allowed but return undefined values.

The returned tensor will have a rank one greater than t. Its outer dimension is the group size and the second dimension will be offsets.dim(1). dims indexes the dimensions of each group in t. This makes it consistent with grouped variants of functions to create tensors. t can be created using createGroupedSliceableTensor() to ensure efficient mapping.

dims and sizes must be size 1.

see multiSlice for informtion on options passed to this function in addition to those passed for plan.

Parameters

graph – The Poplar graph.
t – The tensor being sliced.
offsets – The offsets within t to be sliced.
dims – The dimensions of t for each group.
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

poplar::Tensor multiSlice(poplar::Graph &graph, const poplar::Tensor &t, poplar::ArrayRef<unsigned> offsets, std::size_t dim, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &optionFlags = {})

Take multiple slices from a base tensor.

The returned tensor will have a rank one greater than t. Its outer dimension will be offsets.size(). Note that dim refers to a dimension of t. Any entry in offset greater than equal to the size of dims in t returns a slice filled with zeros.

see other overloads of multiSlice for information on optionFlags.

Parameters

graph – The Poplar graph.
t – The tensor being sliced.
offsets – The offsets within t to be sliced.
dim – The dimension of t to be sliced.
prog – The program to be extended.
debugContext – Optional debug information.
optionFlags – Option flags.

void multiUpdate(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offsets, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Update multiple slices in a tensor.

See overloads of multiSlice for information on options.

dims and sizes must be size 1.

Parameters

graph – The Poplar graph.
t – The tensor being updated.
s – The slices to insert.
offsets – The offsets within t to be updated.
dims – The dimensions of t to be updated.
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

void groupedMultiUpdate(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offsets, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Update multiple slices in a tensor with a group dimension.

The tensors t s , and offsets have group dimension as the first dimension.

dims and sizes must be size 1.

See overloads of multiSlice for information on options.

Parameters

graph – The Poplar graph.
t – The tensor being updated.
s – The slices to insert.
offsets – The offsets within t to be updated.
dims – The dimensions of each group of t to be updated.
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

void multiUpdateAdd(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offsets, const poplar::Tensor &scale, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Accumulate multiple slices in a tensor.

dims and sizes must be size 1.

See overloads of multiSlice for information on options.

Parameters

graph – The Poplar graph.
t – The tensor being updated (must be rank 2).
s – The slices to accumulate.
offsets – The offsets within t to be accumulated.
scale – The scaling to apply to the update. The type of the tensor should be the same as that of t and s except for the case when t and s are of type HALF. In which case scale can be of type FLOAT or HALF.
dims – The dimensions of t to be accumulated (must be rank 1).
sizes – The size of the accumulate in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

void groupedMultiUpdateAdd(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offsets, const poplar::Tensor &scale, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Accumulate multiple slices in a tensor where the first dimension of the tensors t, s and offsets is the group dimension.

t, s must be of the same type

dims and sizes must be size 1.

See overloads of multiSlice for information on options.

Parameters

graph – The Poplar graph.
t – The tensor being updated (must be rank 3).
s – The slices to accumulate.
offsets – The offsets within t to be accumulated.
scale – The scaling to apply to the update. The type of the tensor should be the same as that of t and s except for the case when t and s are of type HALF. In which case scale can be of type FLOAT or HALF.
dims – The dimensions of of each group to be accumulated.
sizes – The size of the accumulate in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

void multiUpdateAdd(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, poplar::ArrayRef<unsigned> offsets, const poplar::Tensor &scale, std::size_t dim, poplar::program::Sequence &prog, const poplar::DebugContext &debugContext = {}, const poplar::OptionFlags &optionFlags = {})

Accumulate multiple slices in a tensor.

t, s must be of the same type

See overloads of multiSlice for information on optionsFlags.

Parameters

graph – The Poplar graph.
t – The tensor being updated.
s – The slices to accumulate.
offsets – The offsets within t to be accumulated.
scale – The scaling to apply to the update. The type of the tensor should be the same as that of t and s except for the case when t and s are of type HALF. In which case scale can be of type FLOAT or HALF.
dim – The dimension of t to be accumulated.
prog – The program to be extended.
debugContext – Optional debug information.
optionFlags – Option flags.

void multiUpdateMax(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offsets, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Find maximum over multiple slices in a tensor.

t, s must have the same element type offsets[i] >= t.dim(0) are ignored.

dims and sizes must be size 1.

See overloads of multiSlice for information on options.

Parameters

graph – The Poplar graph.
t – The tensor being updated (must be rank 2).
s – The slices to find maximum over.
offsets – The offsets within t to find maximum over.
dims – The dimensions of t to find maximum over (must be rank 1).
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

void groupedMultiUpdateMax(poplar::Graph &graph, const poplar::Tensor &t, const poplar::Tensor &s, const poplar::Tensor &offsets, const std::vector<std::size_t> &dims, const std::vector<std::size_t> &sizes, poplar::program::Sequence &prog, const SlicePlan &plan, const poplar::OptionFlags &options, const poplar::DebugContext &debugContext = {})

Find maximum over multiple slices in a tensor with a group dimension.

The tensors t, s, offsets have groups as their first dimension. The offsets tensor contains indices per group which update elements of the corresponding group.

dims and sizes must be size 1.

See overloads of multiSlice for information on options.

Parameters

graph – The Poplar graph.
t – The tensor being updated (must be rank 2).
s – The slices to find maximum over.
offsets – The offsets within t to find maximum over.
dims – The dimensions of each group of t to find maximum over (must be rank 1).
sizes – The size of the update in each of the dimensions in dims.
prog – The program to be extended.
plan – Plan describing how the operation will be implemented.
options – Flags controlling how the operation will be implemented.
debugContext – Optional debug information.

class SlicePlan

#include <DynamicSlice.hpp>

An object representing a plan that describes how to implement a slice or update.

This can be used as a parameter to a function that will slice or update a tensor.

Public Functions

SlicePlan()

~SlicePlan()

SlicePlan(const SlicePlan &other)

SlicePlan(SlicePlan &&other)

SlicePlan &operator=(const SlicePlan &other)

SlicePlan &operator=(SlicePlan &&other)

SlicePlan(std::unique_ptr<SlicePlanInternal> internal)

inline SlicePlanInternal &getImpl() const

Private Members

std::unique_ptr<SlicePlanInternal> internal

Friends

friend std::ostream &operator<<(std::ostream &o, const SlicePlan &p)

friend bool operator<(const SlicePlan &a, const SlicePlan &b) noexcept

friend bool operator==(const SlicePlan &a, const SlicePlan &b) noexcept

friend poplar::ProfileValue toProfileValue(const SlicePlan &p)

namespace embedding

Functions

SlicePlan plan(const poplar::Graph &graph, const poplar::Type &dataType, const std::size_t numEntries, const std::size_t outputSize, const std::vector<std::size_t> &numLookups, const poplar::OptionFlags &options)

Create a plan for implementing a set of operations on an embedding matrix.

** Embedding plan options **

usedForSlice (true, false) [=true]

If true, you intend to use this embedding plan for both a multiSlice operation. An error is thrown if set to false and usedForUpdate is set to false.
usedForUpdate (true, false) [=true]

If true, you intend to use this embedding plan for both a multiUpdate operation. An error is thrown if set to false and usedForSlice is set to false.
operationForUpdate (“none”, “add”, “max”) [=”add”]

Only applicable when usedForUpdate = true. Is the type of operation used in multi-update. Set to “none” for multiUpdate “add” for multiUpdateAdd “max” for multiUpdateMax
availableMemoryProportion Decimal between 0 and 1 (inclusive) [=0.6]

If set, gives the proportion of tile memory made available for temporary variables (variables that become live and die during the operation) for this operation. If not set, the operation has the freedom to use unlimited temporary memory.

See also

createWeights()

See also

Optimising Temporary Memory Usage for Convolutions and Matmuls on the IPU technical note for some practical examples of using availableMemoryProportion
indicesDistribution (uniform, onePoint) [=uniform]

A description of the statistical distribution of the indices that will be sliced/updated over the input size (numEntries) of the operation. This is used to when estimating the runtime of the multiSlice and multiUpdate operation.
- uniform Indices are assumed to be uniformly distributed over the input size of the embedding.
- onePoint Indices are assumed to all be equal.
indicesAreSorted (true, false) [=false]

Plan assuming indices used in MultiUpdate/MultiUpdateOp are sorted in increasing order. The same option must then be used along with the plan when calling MultiUpdate with and without an operation.
validateIndices (true, false) [=false]

Check that all indices are valid at runtime. If any is invalid execution is aborted. Ignored if remapOutOfBoundIndices = true.
partialType (half, float) If not provided, defaults to using the same as the data type. Partials type should always be the same or higher precision than the data type of the embedding matrix. Is applicable only for the case where operationForUpdate is add. It is ignored for all other operations and multi-slice.

Parameters

graph – The graph the operation will be added to.
dataType – The data type of the entries in the embedding matrix and the resulting lookups from the matrix.
numEntries – Input size of embedding matrix.
outputSize – Output size of embedding matrix lookup.
numLookups – Vector of numbers of indices which will be looked up in the embedding matrix.
options – Set of option flags controlling how the operation will be implemented.

Returns

A plan which describes how the embedding matrix lookup/update operations should be implemented.

SlicePlan plan(const poplar::Graph &graph, const poplar::Type &dataType, const std::size_t groupSize, const std::size_t numEntries, const std::size_t outputSize, const std::vector<std::size_t> &numLookups, const poplar::OptionFlags &options): Overload of plan with group size.

std::vector<SlicePlan> planMultiple(const std::vector<SlicePlanningParameters> &spds)

Create multiple plans for implementing a set of operations on an embedding matrix.

The plans are computed in parallel.

Parameters: spds – A vector of parameters to compute slice plans for. For more information about the individual parameters see SlicePlanningParameters and plan.
Returns: A vector of slice plans for each SlicePlanningParameters object passed as input.

struct SlicePlanningParameters

#include <DynamicSlice.hpp>

A complete collection of the parameters required to compute a plan for slice/update operations.

Public Functions

inline SlicePlanningParameters(const poplar::Graph &graph, const poplar::Type &dataType, const std::size_t groupSize, const std::size_t numEntries, const std::size_t outputSize, const std::vector<std::size_t> &numLookups, const poplar::OptionFlags &optionFlags)

inline SlicePlanningParameters(const poplar::Graph &graph, const poplar::Type &dataType, const std::size_t numEntries, const std::size_t outputSize, const std::vector<std::size_t> &numLookups, const poplar::OptionFlags &optionFlags)

Public Members

const poplar::Graph &graph: The graph the operation will be added to.

const poplar::Type dataType: The data type of the entries in the embedding matrix and the resulting lookups from the matrix.

const std::size_t groupSize = 1: The group size.

const std::size_t numEntries: Input size of embedding matrix.

const std::size_t outputSize: Output size of embedding matrix lookup.

const std::vector<std::size_t> numLookups: Vector of numbers of indices which will be looked up in the embedding matrix.

const poplar::OptionFlags optionFlags: Set of option flags controlling how the operation will be implemented.