This is an archive of the discontinued LLVM Phabricator instance.

Differential D109815

[mlir][OpDSL] Update op definitions to make shapes more concise (NFC).
ClosedPublic

Authored by gysit on Sep 15 2021, 1:40 AM.

Details

Reviewers
nicolasvasilache
shabalin
rsuderman
stellaraccident
Commits
rG8f2db36b01c0: [mlir][OpDSL] Update op definitions to make shapes more concise (NFC).
Summary

Express the input shape definitions of convolution and pooling operations in terms of the output shapes, filter shapes, strides, and dilations.

Diff Detail

Event Timeline

gysit created this revision.Sep 15 2021, 1:40 AM
Herald added subscribers: wenzhicui, wrengr, Chia-hungDuan and 20 others. · View Herald TranscriptSep 15 2021, 1:40 AM
gysit requested review of this revision.Sep 15 2021, 1:40 AM
Herald added a reviewer: nicolasvasilache. · View Herald TranscriptSep 15 2021, 1:40 AM
Herald added a project: Restricted Project. · View Herald Transcript
Herald added subscribers: stephenneuendorffer, nicolasvasilache. · View Herald Transcript
gysit added a reviewer: shabalin.Sep 15 2021, 1:43 AM
shabalin accepted this revision.EditedSep 15 2021, 1:57 AM

The new signatures look good to me! They resolve the issue with prior definitions being not precise enough in terms of the dimensions of inputs.

This revision is now accepted and ready to land.Sep 15 2021, 1:57 AM
gysit added reviewers: rsuderman, stellaraccident.Sep 15 2021, 1:59 AM
Harbormaster completed remote builds in B123977: Diff 372657.Sep 15 2021, 2:29 AM
stellaraccident accepted this revision.Sep 15 2021, 7:20 AM
rsuderman accepted this revision.Sep 15 2021, 10:36 AM
Closed by commit rG8f2db36b01c0: [mlir][OpDSL] Update op definitions to make shapes more concise (NFC). (authored by gysit). · Explain WhySep 15 2021, 11:23 PM
This revision was automatically updated to reflect the committed changes.
gysit added a commit: rG8f2db36b01c0: [mlir][OpDSL] Update op definitions to make shapes more concise (NFC)..

Revision Contents

PathSize
mlir/
include/
mlir/
Dialect/
Linalg/
IR/
454 lines
python/
mlir/
dialects/
linalg/
opdsl/
ops/
42 lines

Diff 372872

mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml

Show First 20 Lines • Show All 636 Lines • ▼ Show 20 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2] -> (s0)> shape_map: affine_map<()[s0, s1] -> (s0 + s1)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2] -> (s1)> shape_map: affine_map<()[s0, s1] -> (s1)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2] -> (s2)> shape_map: affine_map<()[s0, s1] -> (s0)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1)[s0, s1, s2] -> (d0 + d1)> - affine_map<(d0, d1)[s0, s1] -> (d0 + d1)>
- affine_map<(d0, d1)[s0, s1, s2] -> (d1)> - affine_map<(d0, d1)[s0, s1] -> (d1)>
- affine_map<(d0, d1)[s0, s1, s2] -> (d0)> - affine_map<(d0, d1)[s0, s1] -> (d0)>
iterator_types: iterator_types:
- parallel - parallel
- reduction - reduction
assignments: assignments:
- !ScalarAssign - !ScalarAssign
arg: O arg: O
value: !ScalarExpression value: !ScalarExpression
scalar_apply: scalar_apply:
Show All 27 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5] -> (s0, s1)> shape_map: affine_map<()[s0, s1, s2, s3] -> (s0 + s1, s2 + s3)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5] -> (s2, s3)> shape_map: affine_map<()[s0, s1, s2, s3] -> (s1, s3)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5] -> (s4, s5)> shape_map: affine_map<()[s0, s1, s2, s3] -> (s0, s2)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3, s4, s5] -> (d0 + d2, d1 + d3)> - affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d0 + d2, d1 + d3)>
- affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3, s4, s5] -> (d2, d3)> - affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d2, d3)>
- affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3, s4, s5] -> (d0, d1)> - affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d0, d1)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
assignments: assignments:
- !ScalarAssign - !ScalarAssign
arg: O arg: O
Show All 29 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8] -> (s0, s1, s2)> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5] -> (s0 + s1, s2 + s3, s4 + s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8] -> (s3, s4, s5)> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5] -> (s1, s3, s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8] -> (s6, s7, s8)> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5] -> (s0, s2, s4)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8] -> (d0 - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5] -> (d0 + d3, d1
+ d3, d1 + d4, d2 + d5)> + d4, d2 + d5)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8] -> (d3, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5] -> (d3, d4, d5)>
d4, d5)> - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5] -> (d0, d1, d2)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8] -> (d0,
d1, d2)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
assignments: assignments:
Show All 31 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7] -> (s0, s1, s2)> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s0, s1 * s2 + s3 * s4,
s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7] -> (s3, s2, s4)> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s3, s5, s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7] -> (s0, s5, s4)> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s0, s1, s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7] -> (s6)> attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s2)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7] -> (s7)> attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6] -> (s4)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6, s7] -> (d0, d1 * - affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6] -> (d0, d1 * s2
s6 + d3 * s7, d4)> + d3 * s4, d4)>
- affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6, s7] -> (d3, d4, - affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6] -> (d3, d4, d2)>
d2)> - affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6] -> (d0, d1, d2)>
- affine_map<(d0, d1, d2, d3, d4)[s0, s1, s2, s3, s4, s5, s6, s7] -> (d0, d1,
d2)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
assignments: assignments:
- !ScalarAssign - !ScalarAssign
Show All 34 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s1, s2, s3)> s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s3,
-> (s4, s5, s3, s6)> s7, s9, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s7, s8, s6)> s1, s5, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s2,
s12] -> (s9, s10)> s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s4,
s12] -> (s11, s12)> s8)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1 * s9 + d4 * s11, d2 * s10 + d5 * s12, d6)> s9, s10] -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d6)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d4, d5, d6, d3)> s9, s10] -> (d4, d5, d6, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1, d2, d3)> s9, s10] -> (d0, d1, d2, d3)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
Show All 37 Lines doc: |-
them to the same data type as the accumulator/output. This includes the zero them to the same data type as the accumulator/output. This includes the zero
point offsets common to quantized operations. point offsets common to quantized operations.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s1, s2, s3)> s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s3,
-> (s4, s5, s3, s6)> s7, s9, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: IZp name: IZp
usage: InputOperand usage: InputOperand
type_var: I32 type_var: I32
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: KZp name: KZp
usage: InputOperand usage: InputOperand
type_var: I32 type_var: I32
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s7, s8, s6)> s1, s5, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s2,
s12] -> (s9, s10)> s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s4,
s12] -> (s11, s12)> s8)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1 * s9 + d4 * s11, d2 * s10 + d5 * s12, d6)> s9, s10] -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d6)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d4, d5, d6, d3)> s9, s10] -> (d4, d5, d6, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> ()> s9, s10] -> ()>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> ()> s9, s10] -> ()>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1, d2, d3)> s9, s10] -> (d0, d1, d2, d3)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
▲ Show 20 Lines • Show All 56 Lines • ▼ Show 20 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s1, s2, s3)> s1, s2 * s3 + s4 * s5, s6 * s7 + s8 * s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s10,
-> (s4, s1, s5, s6)> s1, s4, s8)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s4, s7, s8)> s10, s2, s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s3,
s12] -> (s9, s10)> s7)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s5,
s12] -> (s11, s12)> s9)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d4, d2 * s9 + d5 * s11, d3 * s10 + d6 * s12)> s9, s10] -> (d0, d4, d2 * s3 + d5 * s5, d3 * s7 + d6 * s9)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d1, d4, d5, d6)> s9, s10] -> (d1, d4, d5, d6)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1, d2, d3)> s9, s10] -> (d0, d1, d2, d3)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
Show All 33 Lines doc: |-
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16, s17] -> (s0, s1, s2, s3, s4)> s13, s14] -> (s0, s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9 * s10 + s11 * s12,
s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16, s17] -> (s5, s6, s7, s4, s8)> s13, s14] -> (s3, s7, s11, s13, s14)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16, s17] -> (s0, s9, s10, s11, s8)> s13, s14] -> (s0, s1, s5, s9, s14)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16, s17] -> (s12, s13, s14)> s12, s13, s14] -> (s2, s6, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16, s17] -> (s15, s16, s17)> s12, s13, s14] -> (s4, s8, s12)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6,
s7, s8, s9, s10, s11, s12, s13, s14, s15, s16, s17] -> (d0, d1 * s12 + d5 * s7, s8, s9, s10, s11, s12, s13, s14] -> (d0, d1 * s2 + d5 * s4, d2 * s6 + d6
s15, d2 * s13 + d6 * s16, d3 * s14 + d7 * s17, d8)> * s8, d3 * s10 + d7 * s12, d8)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6,
s7, s8, s9, s10, s11, s12, s13, s14, s15, s16, s17] -> (d5, d6, d7, d8, d4)> s7, s8, s9, s10, s11, s12, s13, s14] -> (d5, d6, d7, d8, d4)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7, d8)[s0, s1, s2, s3, s4, s5, s6,
s7, s8, s9, s10, s11, s12, s13, s14, s15, s16, s17] -> (d0, d1, d2, d3, d4)> s7, s8, s9, s10, s11, s12, s13, s14] -> (d0, d1, d2, d3, d4)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
Show All 35 Lines doc: |-
them to the same data type as the accumulator/output. Multiplier is set to 1 them to the same data type as the accumulator/output. Multiplier is set to 1
which is a special case for most dpethwise convolutions. which is a special case for most dpethwise convolutions.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1 *
(s0, s1, s2, s3)> s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s3, s7, s9)>
(s4, s5, s3)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s5,
(s0, s6, s7, s3)> s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s2, s6)>
-> (s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s4, s8)>
-> (s10, s11)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1 * s8 + d4 * s10, d2 * s9 + d5 * s11, d3)> -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d4, d5, d3)> -> (d4, d5, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2, d3)> -> (d0, d1, d2, d3)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
assignments: assignments:
Show All 31 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1 *
(s0, s1, s2, s3)> s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s3, s7, s9)>
(s4, s5, s3)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: IZp name: IZp
usage: InputOperand usage: InputOperand
type_var: I32 type_var: I32
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: KZp name: KZp
usage: InputOperand usage: InputOperand
type_var: I32 type_var: I32
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s5,
(s0, s6, s7, s3)> s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s2, s6)>
-> (s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s4, s8)>
-> (s10, s11)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1 * s8 + d4 * s10, d2 * s9 + d5 * s11, d3)> -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d4, d5, d3)> -> (d4, d5, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> ()> -> ()>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> ()> -> ()>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2, d3)> -> (d0, d1, d2, d3)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
assignments: assignments:
▲ Show 20 Lines • Show All 51 Lines • ▼ Show 20 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s1, s2, s3)> s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s3,
-> (s4, s5, s3, s6)> s7, s9, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s7, s8, s3, s6)> s1, s5, s9, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s2,
s12] -> (s9, s10)> s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s4,
s12] -> (s11, s12)> s8)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1 * s9 + d5 * s11, d2 * s10 + d6 * s12, d3)> s9, s10] -> (d0, d1 * s2 + d5 * s4, d2 * s6 + d6 * s8, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d5, d6, d3, d4)> s9, s10] -> (d5, d6, d3, d4)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1, d2, d3, d4)> s9, s10] -> (d0, d1, d2, d3, d4)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
Show All 32 Lines doc: |-
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s1, s2, s3)> s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s3,
-> (s4, s5, s3, s6)> s7, s9, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: IZp name: IZp
usage: InputOperand usage: InputOperand
type_var: I32 type_var: I32
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: KZp name: KZp
usage: InputOperand usage: InputOperand
type_var: I32 type_var: I32
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12] shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s0,
-> (s0, s7, s8, s3, s6)> s1, s5, s9, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s2,
s12] -> (s9, s10)> s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10] -> (s4,
s12] -> (s11, s12)> s8)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1 * s9 + d5 * s11, d2 * s10 + d6 * s12, d3)> s9, s10] -> (d0, d1 * s2 + d5 * s4, d2 * s6 + d6 * s8, d3)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d5, d6, d3, d4)> s9, s10] -> (d5, d6, d3, d4)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> ()> s9, s10] -> ()>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> ()> s9, s10] -> ()>
- affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8, - affine_map<(d0, d1, d2, d3, d4, d5, d6)[s0, s1, s2, s3, s4, s5, s6, s7, s8,
s9, s10, s11, s12] -> (d0, d1, d2, d3, d4)> s9, s10] -> (d0, d1, d2, d3, d4)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
▲ Show 20 Lines • Show All 52 Lines • ▼ Show 20 Lines doc: |-
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1 *
(s0, s1, s2, s3)> s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s3, s7)>
(s4, s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s5,
(s0, s6, s7, s3)> s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s2, s6)>
-> (s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s4, s8)>
-> (s10, s11)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1 * s8 + d3 * s10, d2 * s9 + d4 * s11, d5)> -> (d0, d1 * s2 + d3 * s4, d2 * s6 + d4 * s8, d5)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d3, d4)> -> (d3, d4)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2, d5)> -> (d0, d1, d2, d5)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- parallel - parallel
assignments: assignments:
Show All 21 Lines doc: |-
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1 *
(s0, s1, s2, s3)> s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s3, s7)>
(s4, s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s5,
(s0, s6, s7, s3)> s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s2, s6)>
-> (s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s4, s8)>
-> (s10, s11)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1 * s8 + d3 * s10, d2 * s9 + d4 * s11, d5)> -> (d0, d1 * s2 + d3 * s4, d2 * s6 + d4 * s8, d5)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d3, d4)> -> (d3, d4)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2, d5)> -> (d0, d1, d2, d5)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- parallel - parallel
assignments: assignments:
Show All 21 Lines doc: |-
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s2
(s0, s1, s2, s3)> * s3 + s4 * s5, s6 * s7 + s8 * s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s4, s8)>
(s4, s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s2,
(s0, s1, s6, s7)> s6)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s3, s7)>
-> (s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s5, s9)>
-> (s10, s11)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2 * s8 + d4 * s10, d3 * s9 + d5 * s11)> -> (d0, d1, d2 * s3 + d4 * s5, d3 * s7 + d5 * s9)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d4, d5)> -> (d4, d5)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2, d3)> -> (d0, d1, d2, d3)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
assignments: assignments:
Show All 21 Lines doc: |-
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1 *
(s0, s1, s2, s3)> s2 + s3 * s4, s5 * s6 + s7 * s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s3, s7)>
(s4, s5)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] -> shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s0, s1, s5,
(s0, s6, s7, s3)> s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s2, s6)>
-> (s8, s9)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11] attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9] -> (s4, s8)>
-> (s10, s11)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1 * s8 + d3 * s10, d2 * s9 + d4 * s11, d5)> -> (d0, d1 * s2 + d3 * s4, d2 * s6 + d4 * s8, d5)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d3, d4)> -> (d3, d4)>
- affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, - affine_map<(d0, d1, d2, d3, d4, d5)[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9]
s10, s11] -> (d0, d1, d2, d5)> -> (d0, d1, d2, d5)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- parallel - parallel
assignments: assignments:
Show All 22 Lines doc: |-
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s0, s1, s2, s3, s4)> s13] -> (s0, s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9 * s10 + s11 * s12, s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s5, s6, s7)> s13] -> (s3, s7, s11)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s0, s8, s9, s10, s4)> s13] -> (s0, s1, s5, s9, s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16] -> (s11, s12, s13)> s12, s13] -> (s2, s6, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16] -> (s14, s15, s16)> s12, s13] -> (s4, s8, s12)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d0, d1 * s11 + d4 * s14, d2 * s8, s9, s10, s11, s12, s13] -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d3
s12 + d5 * s15, d3 * s13 + d6 * s16, d7)> * s10 + d6 * s12, d7)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d4, d5, d6)> s8, s9, s10, s11, s12, s13] -> (d4, d5, d6)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d0, d1, d2, d3, d7)> s8, s9, s10, s11, s12, s13] -> (d0, d1, d2, d3, d7)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
Show All 24 Lines doc: |-
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s0, s1, s2, s3, s4)> s13] -> (s0, s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9 * s10 + s11 * s12, s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s5, s6, s7)> s13] -> (s3, s7, s11)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s0, s8, s9, s10, s4)> s13] -> (s0, s1, s5, s9, s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16] -> (s11, s12, s13)> s12, s13] -> (s2, s6, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16] -> (s14, s15, s16)> s12, s13] -> (s4, s8, s12)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d0, d1 * s11 + d4 * s14, d2 * s8, s9, s10, s11, s12, s13] -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d3
s12 + d5 * s15, d3 * s13 + d6 * s16, d7)> * s10 + d6 * s12, d7)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d4, d5, d6)> s8, s9, s10, s11, s12, s13] -> (d4, d5, d6)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d0, d1, d2, d3, d7)> s8, s9, s10, s11, s12, s13] -> (d0, d1, d2, d3, d7)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
Show All 24 Lines doc: |-
data type as the accumulator/output. data type as the accumulator/output.
structured_op: !LinalgStructuredOpConfig structured_op: !LinalgStructuredOpConfig
args: args:
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: I name: I
usage: InputOperand usage: InputOperand
type_var: T1 type_var: T1
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s0, s1, s2, s3, s4)> s13] -> (s0, s1 * s2 + s3 * s4, s5 * s6 + s7 * s8, s9 * s10 + s11 * s12, s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: K name: K
usage: InputOperand usage: InputOperand
type_var: T2 type_var: T2
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s5, s6, s7)> s13] -> (s3, s7, s11)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: O name: O
usage: OutputOperand usage: OutputOperand
type_var: U type_var: U
shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12, shape_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, s12,
s13, s14, s15, s16] -> (s0, s8, s9, s10, s4)> s13] -> (s0, s1, s5, s9, s13)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: strides name: strides
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16] -> (s11, s12, s13)> s12, s13] -> (s2, s6, s10)>
- !LinalgOperandDefConfig - !LinalgOperandDefConfig
name: dilations name: dilations
usage: IndexAttribute usage: IndexAttribute
type_var: I64 type_var: I64
attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11, attribute_map: affine_map<()[s0, s1, s2, s3, s4, s5, s6, s7, s8, s9, s10, s11,
s12, s13, s14, s15, s16] -> (s14, s15, s16)> s12, s13] -> (s4, s8, s12)>
indexing_maps: !LinalgIndexingMapsConfig indexing_maps: !LinalgIndexingMapsConfig
static_indexing_maps: static_indexing_maps:
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d0, d1 * s11 + d4 * s14, d2 * s8, s9, s10, s11, s12, s13] -> (d0, d1 * s2 + d4 * s4, d2 * s6 + d5 * s8, d3
s12 + d5 * s15, d3 * s13 + d6 * s16, d7)> * s10 + d6 * s12, d7)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d4, d5, d6)> s8, s9, s10, s11, s12, s13] -> (d4, d5, d6)>
- affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7, - affine_map<(d0, d1, d2, d3, d4, d5, d6, d7)[s0, s1, s2, s3, s4, s5, s6, s7,
s8, s9, s10, s11, s12, s13, s14, s15, s16] -> (d0, d1, d2, d3, d7)> s8, s9, s10, s11, s12, s13] -> (d0, d1, d2, d3, d7)>
iterator_types: iterator_types:
- parallel - parallel
- parallel - parallel
- parallel - parallel
- parallel - parallel
- reduction - reduction
- reduction - reduction
- reduction - reduction
▲ Show 20 Lines • Show All 224 Lines • Show Last 20 Lines

mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py

Show First 20 Lines • Show All 140 Lines • ▼ Show 20 Linesdef dot(
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
implements(ContractionOpInterface) implements(ContractionOpInterface)
C[None] += cast(U, A[D.m]) * cast(U, B[D.m]) C[None] += cast(U, A[D.m]) * cast(U, B[D.m])
@linalg_structured_op @linalg_structured_op
def conv_1d( def conv_1d(
I=TensorDef(T1, S.IW), I=TensorDef(T1, S.OW + S.KW),
K=TensorDef(T2, S.KW), K=TensorDef(T2, S.KW),
O=TensorDef(U, S.OW, output=True)): O=TensorDef(U, S.OW, output=True)):
"""Performs 1-D convolution with no channels. """Performs 1-D convolution with no channels.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.ow, D.kw) domain(D.ow, D.kw)
O[D.ow] += cast( O[D.ow] += cast(
U, I[D.ow + D.kw]) * cast(U, K[D.kw]) U, I[D.ow + D.kw]) * cast(U, K[D.kw])
@linalg_structured_op @linalg_structured_op
def conv_2d( def conv_2d(
I=TensorDef(T1, S.IH, S.IW), I=TensorDef(T1, S.OH + S.KH, S.OW + S.KW),
K=TensorDef(T2, S.KH, S.KW), K=TensorDef(T2, S.KH, S.KW),
O=TensorDef(U, S.OH, S.OW, output=True)): O=TensorDef(U, S.OH, S.OW, output=True)):
"""Performs 2-D convolution with no channels. """Performs 2-D convolution with no channels.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.oh, D.ow, D.kh, D.kw) domain(D.oh, D.ow, D.kh, D.kw)
O[D.oh, D.ow] += cast( O[D.oh, D.ow] += cast(
U, I[D.oh + D.kh, D.ow + D.kw]) * cast(U, K[D.kh, D.kw]) U, I[D.oh + D.kh, D.ow + D.kw]) * cast(U, K[D.kh, D.kw])
@linalg_structured_op @linalg_structured_op
def conv_3d( def conv_3d(
I=TensorDef(T1, S.ID, S.IH, S.IW), I=TensorDef(T1, S.OD + S.KD, S.OH + S.KH, S.OW + S.KW),
K=TensorDef(T2, S.KD, S.KH, S.KW), K=TensorDef(T2, S.KD, S.KH, S.KW),
O=TensorDef(U, S.OD, S.OH, S.OW, output=True)): O=TensorDef(U, S.OD, S.OH, S.OW, output=True)):
"""Performs 3-D convolution with no channels. """Performs 3-D convolution with no channels.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.od, D.oh, D.ow, D.kd, D.kh, D.kw) domain(D.od, D.oh, D.ow, D.kd, D.kh, D.kw)
O[D.od, D.oh, D.ow] += cast( O[D.od, D.oh, D.ow] += cast(
U, I[D.od + D.kd, D.oh + D.kh, D.ow + D.kw]) * cast(U, K[D.kd, D.kh, D.kw]) U, I[D.od + D.kd, D.oh + D.kh, D.ow + D.kw]) * cast(U, K[D.kd, D.kh, D.kw])
@linalg_structured_op @linalg_structured_op
def conv_1d_nwc_wcf( def conv_1d_nwc_wcf(
I=TensorDef(T1, S.N, S.IW, S.C), I=TensorDef(T1, S.N, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KW, S.C, S.F), K=TensorDef(T2, S.KW, S.C, S.F),
O=TensorDef(U, S.N, S.OW, S.F, output=True), O=TensorDef(U, S.N, S.OW, S.F, output=True),
strides=AttributeDef(S.SW), strides=AttributeDef(S.SW),
dilations=AttributeDef(S.DW)): dilations=AttributeDef(S.DW)):
"""Performs 1-D convolution. """Performs 1-D convolution.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.ow, D.f, D.kw, D.c) domain(D.n, D.ow, D.f, D.kw, D.c)
O[D.n, D.ow, D.f] += cast( O[D.n, D.ow, D.f] += cast(
U, I[D.n, D.ow * S.SW + D.kw * S.DW, D.c U, I[D.n, D.ow * S.SW + D.kw * S.DW, D.c
]) * cast(U, K[D.kw, D.c, D.f]) ]) * cast(U, K[D.kw, D.c, D.f])
@linalg_structured_op @linalg_structured_op
def conv_2d_nhwc_hwcf( def conv_2d_nhwc_hwcf(
I=TensorDef(T1, S.N, S.IH, S.IW, S.C), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KH, S.KW, S.C, S.F), K=TensorDef(T2, S.KH, S.KW, S.C, S.F),
O=TensorDef(U, S.N, S.OH, S.OW, S.F, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.F, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs 2-D convolution. """Performs 2-D convolution.
Layout: Layout:
* Input: NHWC. * Input: NHWC.
* Kernel: HWCF. * Kernel: HWCF.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.f, D.kh, D.kw, D.c) domain(D.n, D.oh, D.ow, D.f, D.kh, D.kw, D.c)
O[D.n, D.oh, D.ow, D.f] += cast( O[D.n, D.oh, D.ow, D.f] += cast(
U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c
]) * cast(U, K[D.kh, D.kw, D.c, D.f]) ]) * cast(U, K[D.kh, D.kw, D.c, D.f])
@linalg_structured_op @linalg_structured_op
def conv_2d_nhwc_hwcf_q( def conv_2d_nhwc_hwcf_q(
I=TensorDef(T1, S.N, S.IH, S.IW, S.C), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KH, S.KW, S.C, S.F), K=TensorDef(T2, S.KH, S.KW, S.C, S.F),
IZp=ScalarDef(I32), IZp=ScalarDef(I32),
KZp=ScalarDef(I32), KZp=ScalarDef(I32),
O=TensorDef(U, S.N, S.OH, S.OW, S.F, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.F, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs 2-D convolution with zero point offsets. """Performs 2-D convolution with zero point offsets.
Layout: Layout:
* Input: NHWC. * Input: NHWC.
* Kernel: HWCF. * Kernel: HWCF.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. This includes the zero them to the same data type as the accumulator/output. This includes the zero
point offsets common to quantized operations. point offsets common to quantized operations.
""" """
domain(D.n, D.oh, D.ow, D.f, D.kh, D.kw, D.c) domain(D.n, D.oh, D.ow, D.f, D.kh, D.kw, D.c)
O[D.n, D.oh, D.ow, D.f] += (cast( O[D.n, D.oh, D.ow, D.f] += (cast(
U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c
]) - cast(U, IZp)) * (cast(U, K[D.kh, D.kw, D.c, D.f]) - cast(U, KZp)) ]) - cast(U, IZp)) * (cast(U, K[D.kh, D.kw, D.c, D.f]) - cast(U, KZp))
@linalg_structured_op @linalg_structured_op
def conv_2d_nchw_fchw( def conv_2d_nchw_fchw(
I=TensorDef(T1, S.N, S.C, S.IH, S.IW), I=TensorDef(T1, S.N, S.C, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW),
K=TensorDef(T2, S.F, S.C, S.KH, S.KW), K=TensorDef(T2, S.F, S.C, S.KH, S.KW),
O=TensorDef(U, S.N, S.F, S.OH, S.OW, output=True), O=TensorDef(U, S.N, S.F, S.OH, S.OW, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs 2-D convolution. """Performs 2-D convolution.
Layout: Layout:
* Input: NCHW. * Input: NCHW.
* Kernel: FCHW. * Kernel: FCHW.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.f, D.oh, D.ow, D.c, D.kh, D.kw) domain(D.n, D.f, D.oh, D.ow, D.c, D.kh, D.kw)
O[D.n, D.f, D.oh, D.ow] += cast( O[D.n, D.f, D.oh, D.ow] += cast(
U, I[D.n, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW U, I[D.n, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW
]) * cast(U, K[D.f, D.c, D.kh, D.kw]) ]) * cast(U, K[D.f, D.c, D.kh, D.kw])
@linalg_structured_op @linalg_structured_op
def conv_3d_ndhwc_dhwcf( def conv_3d_ndhwc_dhwcf(
I=TensorDef(T1, S.N, S.ID, S.IH, S.IW, S.C), I=TensorDef(T1, S.N, S.OD * S.SD + S.KD * S.DD,
S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KD, S.KH, S.KW, S.C, S.F), K=TensorDef(T2, S.KD, S.KH, S.KW, S.C, S.F),
O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.F, output=True), O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.F, output=True),
strides=AttributeDef(S.SD, S.SH, S.SW), strides=AttributeDef(S.SD, S.SH, S.SW),
dilations=AttributeDef(S.DD, S.DH, S.DW)): dilations=AttributeDef(S.DD, S.DH, S.DW)):
"""Performs 3-D convolution. """Performs 3-D convolution.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.od, D.oh, D.ow, D.f, D.kd, D.kh, D.kw, D.c) domain(D.n, D.od, D.oh, D.ow, D.f, D.kd, D.kh, D.kw, D.c)
O[D.n, D.od, D.oh, D.ow, D.f] += cast( O[D.n, D.od, D.oh, D.ow, D.f] += cast(
U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c
]) * cast(U, K[D.kd, D.kh, D.kw, D.c, D.f]) ]) * cast(U, K[D.kd, D.kh, D.kw, D.c, D.f])
@linalg_structured_op @linalg_structured_op
def depthwise_conv2D_nhw( def depthwise_conv2D_nhw(
I=TensorDef(T1, S.N, S.IH, S.IW, S.IC), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.IC),
K=TensorDef(T2, S.KH, S.KW, S.IC), K=TensorDef(T2, S.KH, S.KW, S.IC),
O=TensorDef(U, S.N, S.OH, S.OW, S.IC, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.IC, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs depth-wise 2-D convolution. """Performs depth-wise 2-D convolution.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. Multiplier is set to 1 them to the same data type as the accumulator/output. Multiplier is set to 1
which is a special case for most dpethwise convolutions. which is a special case for most dpethwise convolutions.
""" """
domain(D.n, D.oh, D.ow, D.ic, D.kh, D.kw) domain(D.n, D.oh, D.ow, D.ic, D.kh, D.kw)
O[D.n, D.oh, D.ow, D.ic] += cast( O[D.n, D.oh, D.ow, D.ic] += cast(
U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
D.ic]) * cast(U, K[D.kh, D.kw, D.ic]) D.ic]) * cast(U, K[D.kh, D.kw, D.ic])
@linalg_structured_op @linalg_structured_op
def depthwise_conv2D_nhw_q( def depthwise_conv2D_nhw_q(
I=TensorDef(T1, S.N, S.IH, S.IW, S.IC), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.IC),
K=TensorDef(T2, S.KH, S.KW, S.IC), K=TensorDef(T2, S.KH, S.KW, S.IC),
IZp=ScalarDef(I32), IZp=ScalarDef(I32),
KZp=ScalarDef(I32), KZp=ScalarDef(I32),
O=TensorDef(U, S.N, S.OH, S.OW, S.IC, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.IC, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs depth-wise 2-D convolution. """Performs depth-wise 2-D convolution.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.ic, D.kh, D.kw) domain(D.n, D.oh, D.ow, D.ic, D.kh, D.kw)
O[D.n, D.oh, D.ow, D.ic] += ( O[D.n, D.oh, D.ow, D.ic] += (
(cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, (cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
D.ic]) - cast(U, IZp)) * D.ic]) - cast(U, IZp)) *
(cast(U, K[D.kh, D.kw, D.ic]) - cast(U, KZp))) (cast(U, K[D.kh, D.kw, D.ic]) - cast(U, KZp)))
@linalg_structured_op @linalg_structured_op
def depthwise_conv2D_nhwc( def depthwise_conv2D_nhwc(
I=TensorDef(T1, S.N, S.IH, S.IW, S.IC), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.IC),
K=TensorDef(T2, S.KH, S.KW, S.IC, S.CM), K=TensorDef(T2, S.KH, S.KW, S.IC, S.CM),
O=TensorDef(U, S.N, S.OH, S.OW, S.IC, S.CM, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.IC, S.CM, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs depth-wise 2-D convolution. """Performs depth-wise 2-D convolution.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.ic, D.cm, D.kh, D.kw) domain(D.n, D.oh, D.ow, D.ic, D.cm, D.kh, D.kw)
O[D.n, D.oh, D.ow, D.ic, D.cm] += cast( O[D.n, D.oh, D.ow, D.ic, D.cm] += cast(
U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
D.ic]) * cast(U, K[D.kh, D.kw, D.ic, D.cm]) D.ic]) * cast(U, K[D.kh, D.kw, D.ic, D.cm])
@linalg_structured_op @linalg_structured_op
def depthwise_conv2D_nhwc_q( def depthwise_conv2D_nhwc_q(
I=TensorDef(T1, S.N, S.IH, S.IW, S.IC), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.IC),
K=TensorDef(T2, S.KH, S.KW, S.IC, S.CM), K=TensorDef(T2, S.KH, S.KW, S.IC, S.CM),
IZp=ScalarDef(I32), IZp=ScalarDef(I32),
KZp=ScalarDef(I32), KZp=ScalarDef(I32),
O=TensorDef(U, S.N, S.OH, S.OW, S.IC, S.CM, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.IC, S.CM, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs depth-wise 2-D convolution. """Performs depth-wise 2-D convolution.
Numeric casting is performed on the operands to the inner multiply, promoting Numeric casting is performed on the operands to the inner multiply, promoting
them to the same data type as the accumulator/output. them to the same data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.ic, D.cm, D.kh, D.kw) domain(D.n, D.oh, D.ow, D.ic, D.cm, D.kh, D.kw)
O[D.n, D.oh, D.ow, D.ic, D.cm] += ( O[D.n, D.oh, D.ow, D.ic, D.cm] += (
(cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, (cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
D.ic]) - cast(U, IZp)) * D.ic]) - cast(U, IZp)) *
(cast(U, K[D.kh, D.kw, D.ic, D.cm]) - cast(U, KZp))) (cast(U, K[D.kh, D.kw, D.ic, D.cm]) - cast(U, KZp)))
@linalg_structured_op @linalg_structured_op
def pooling_nhwc_sum( def pooling_nhwc_sum(
I=TensorDef(T1, S.N, S.H, S.W, S.C), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]), K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]),
O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs sum pooling. """Performs sum pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.kh, D.kw, D.c) domain(D.n, D.oh, D.ow, D.kh, D.kw, D.c)
O[D.n, D.oh, D.ow, D.c] += cast( O[D.n, D.oh, D.ow, D.c] += cast(
U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c]) U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, D.c])
@linalg_structured_op @linalg_structured_op
def pooling_nhwc_max( def pooling_nhwc_max(
I=TensorDef(T1, S.N, S.H, S.W, S.C), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]), K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]),
O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs max pooling. """Performs max pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.kh, D.kw, D.c) domain(D.n, D.oh, D.ow, D.kh, D.kw, D.c)
O[D.n, D.oh, D.ow, D.c] = ReduceFn.max(D.kh, D.kw)( O[D.n, D.oh, D.ow, D.c] = ReduceFn.max(D.kh, D.kw)(
cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
D.c])) D.c]))
@linalg_structured_op @linalg_structured_op
def pooling_nchw_max( def pooling_nchw_max(
I=TensorDef(T1, S.N, S.C, S.H, S.W), I=TensorDef(T1, S.N, S.C, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW),
K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]), K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]),
O=TensorDef(U, S.N, S.C, S.OH, S.OW, output=True), O=TensorDef(U, S.N, S.C, S.OH, S.OW, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs max pooling. """Performs max pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
""" """
domain(D.n, D.c, D.oh, D.ow, D.kh, D.kw) domain(D.n, D.c, D.oh, D.ow, D.kh, D.kw)
O[D.n, D.c, D.oh, D.ow] = ReduceFn.max(D.kh, D.kw)( O[D.n, D.c, D.oh, D.ow] = ReduceFn.max(D.kh, D.kw)(
cast(U, I[D.n, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, cast(U, I[D.n, D.c, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
])) ]))
@linalg_structured_op @linalg_structured_op
def pooling_nhwc_min( def pooling_nhwc_min(
I=TensorDef(T1, S.N, S.H, S.W, S.C), I=TensorDef(T1, S.N, S.OH * S.SH + S.KH * S.DH, S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]), K=TensorDef(T2, S.KH, S.KW, index_dims=[D.kh, D.kw]),
O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True), O=TensorDef(U, S.N, S.OH, S.OW, S.C, output=True),
strides=AttributeDef(S.SH, S.SW), strides=AttributeDef(S.SH, S.SW),
dilations=AttributeDef(S.DH, S.DW)): dilations=AttributeDef(S.DH, S.DW)):
"""Performs min pooling. """Performs min pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
""" """
domain(D.n, D.oh, D.ow, D.kh, D.kw, D.c) domain(D.n, D.oh, D.ow, D.kh, D.kw, D.c)
O[D.n, D.oh, D.ow, D.c] = ReduceFn.min(D.kh, D.kw)( O[D.n, D.oh, D.ow, D.c] = ReduceFn.min(D.kh, D.kw)(
cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW, cast(U, I[D.n, D.oh * S.SH + D.kh * S.DH, D.ow * S.SW + D.kw * S.DW,
D.c])) D.c]))
@linalg_structured_op @linalg_structured_op
def pooling_ndhwc_sum( def pooling_ndhwc_sum(
I=TensorDef(T1, S.N, S.D, S.H, S.W, S.C), I=TensorDef(T1, S.N, S.OD * S.SD + S.KD * S.DD, S.OH * S.SH + S.KH * S.DH,
S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KD, S.KH, S.KW, index_dims=[D.kd, D.kh, D.kw]), K=TensorDef(T2, S.KD, S.KH, S.KW, index_dims=[D.kd, D.kh, D.kw]),
O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.C, output=True), O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.C, output=True),
strides=AttributeDef(S.SD, S.SH, S.SW), strides=AttributeDef(S.SD, S.SH, S.SW),
dilations=AttributeDef(S.DD, S.DH, S.DW)): dilations=AttributeDef(S.DD, S.DH, S.DW)):
"""Performs 3D sum pooling. """Performs 3D sum pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
""" """
domain(D.n, D.od, D.oh, D.ow, D.kd, D.kh, D.kw, D.c) domain(D.n, D.od, D.oh, D.ow, D.kd, D.kh, D.kw, D.c)
O[D.n, D.od, D.oh, D.ow, D.c] += cast( O[D.n, D.od, D.oh, D.ow, D.c] += cast(
U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH, U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH,
D.ow * S.SW + D.kw * S.DW, D.c]) D.ow * S.SW + D.kw * S.DW, D.c])
@linalg_structured_op @linalg_structured_op
def pooling_ndhwc_max( def pooling_ndhwc_max(
I=TensorDef(T1, S.N, S.D, S.H, S.W, S.C), I=TensorDef(T1, S.N, S.OD * S.SD + S.KD * S.DD, S.OH * S.SH + S.KH * S.DH,
S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KD, S.KH, S.KW, index_dims=[D.kd, D.kh, D.kw]), K=TensorDef(T2, S.KD, S.KH, S.KW, index_dims=[D.kd, D.kh, D.kw]),
O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.C, output=True), O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.C, output=True),
strides=AttributeDef(S.SD, S.SH, S.SW), strides=AttributeDef(S.SD, S.SH, S.SW),
dilations=AttributeDef(S.DD, S.DH, S.DW)): dilations=AttributeDef(S.DD, S.DH, S.DW)):
"""Performs 3D max pooling. """Performs 3D max pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
""" """
domain(D.n, D.od, D.oh, D.ow, D.kd, D.kh, D.kw, D.c) domain(D.n, D.od, D.oh, D.ow, D.kd, D.kh, D.kw, D.c)
O[D.n, D.od, D.oh, D.ow, D.c] = ReduceFn.max(D.kd, D.kh, D.kw)( O[D.n, D.od, D.oh, D.ow, D.c] = ReduceFn.max(D.kd, D.kh, D.kw)(
cast( cast(
U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH, U, I[D.n, D.od * S.SD + D.kd * S.DD, D.oh * S.SH + D.kh * S.DH,
D.ow * S.SW + D.kw * S.DW, D.c])) D.ow * S.SW + D.kw * S.DW, D.c]))
@linalg_structured_op @linalg_structured_op
def pooling_ndhwc_min( def pooling_ndhwc_min(
I=TensorDef(T1, S.N, S.D, S.H, S.W, S.C), I=TensorDef(T1, S.N, S.OD * S.SD + S.KD * S.DD, S.OH * S.SH + S.KH * S.DH,
S.OW * S.SW + S.KW * S.DW, S.C),
K=TensorDef(T2, S.KD, S.KH, S.KW, index_dims=[D.kd, D.kh, D.kw]), K=TensorDef(T2, S.KD, S.KH, S.KW, index_dims=[D.kd, D.kh, D.kw]),
O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.C, output=True), O=TensorDef(U, S.N, S.OD, S.OH, S.OW, S.C, output=True),
strides=AttributeDef(S.SD, S.SH, S.SW), strides=AttributeDef(S.SD, S.SH, S.SW),
dilations=AttributeDef(S.DD, S.DH, S.DW)): dilations=AttributeDef(S.DD, S.DH, S.DW)):
"""Performs 3D min pooling. """Performs 3D min pooling.
Numeric casting is performed on the input operand, promoting it to the same Numeric casting is performed on the input operand, promoting it to the same
data type as the accumulator/output. data type as the accumulator/output.
▲ Show 20 Lines • Show All 46 Lines • Show Last 20 Lines