Home
last modified time | relevance | path

Searched refs:broadcasted (Results 1 – 25 of 28) sorted by relevance

12

/external/tensorflow/tensorflow/core/api_def/base_api/
Dapi_def_BroadcastTo.pbtxt40 is broadcasted to output Tensor with shape of `[3, 3]`.
42 When doing broadcasted operations such as multiplying a tensor
44 benefit, as the broadcasted tensor is never materialized.
47 The newly-created tensor takes the full memory of the broadcasted
Dapi_def_BroadcastArgs.pbtxt9 broadcasted shape. `s0`, `s1` and `r0` are all integer vectors.
Dapi_def_Einsum.pbtxt79 named axis label. If two inputs contain ellipsis, then they are broadcasted
83 The broadcasted dimensions are placed in the corresponding location of the
84 ellipsis in the output subscript. If the broadcasted dimensions are non-empty
/external/tensorflow/tensorflow/compiler/mlir/tools/kernel_gen/ir/
Dtf_framework_ops.td176 let summary = "Minimizes the rank of two or more shapes to be broadcasted";
182 reshape of the corresponding input shape, and the broadcasted output shape
184 broadcasted output shape of the input shapes. Among all possibilities with
201 The broadcasted output shape of the operands is [1, 2, 3, 1, 2, 3], the
202 broadcasted output shape of the outputs is [6, 2, 3]. These two shapes are
/external/tensorflow/tensorflow/compiler/mlir/tfjs/tests/
Dops.mlir14 …// expected-error @+1 {{op result type '1x2x3x5' not broadcast compatible with broadcasted operand…
22 …// expected-error @+1 {{result type '7x3x2x14' not broadcast compatible with broadcasted operands'…
/external/llvm-project/mlir/test/Dialect/
Dtraits.mlir81 …// expected-error @+1 {{op result type '4x3x3' not broadcast compatible with broadcasted operands'…
91 …// expected-error @+1 {{op result type '8x7x6x1' not broadcast compatible with broadcasted operand…
134 …// expected-error @+1 {{op result type '2' not broadcast compatible with broadcasted operands's sh…
/external/mesa3d/docs/relnotes/
D19.2.7.rst36 vertex's attribute to be broadcasted for flat interpolation
/external/autotest/server/site_tests/network_WiFi_DarkResumeActiveScans/
Dcontrol19 (e.g. MAC addresses) to be broadcasted.
/external/llvm-project/mlir/lib/Conversion/SPIRVToLLVM/
DConvertSPIRVToLLVM.cpp145 Value broadcasted = rewriter.create<LLVM::UndefOp>(loc, llvmVectorType); in broadcast() local
149 broadcasted = rewriter.create<LLVM::InsertElementOp>( in broadcast()
150 loc, llvmVectorType, broadcasted, toBroadcast, index); in broadcast()
152 return broadcasted; in broadcast()
179 Value broadcasted = in processCountOrOffset() local
181 return optionallyTruncateOrExtend(loc, broadcasted, dstType, rewriter); in processCountOrOffset()
/external/tensorflow/tensorflow/compiler/xla/tests/
Dreduce_test.cc566 auto broadcasted = Broadcast(scalar, {500, 500}); in XLA_TEST_F() local
567 Reduce(broadcasted, ConstantR0<float>(&builder, 0.0f), add, {0, 1}); in XLA_TEST_F()
578 auto broadcasted = Broadcast(scalar, {500, 500}); in XLA_TEST_F() local
579 Reduce(broadcasted, ConstantR0<float>(&builder, 0.0f), max, {0, 1}); in XLA_TEST_F()
/external/llvm-project/mlir/include/mlir/Dialect/Shape/IR/
DShapeOps.td51 let summary = "Returns the broadcasted output shape of two inputs";
53 Returns the broadcasted shape for two input shapes or extent tensors. Both
59 with 1's from the left. The resulting broadcasted shape is then defined as
187 let summary = "Determines if 2 shapes can be successfully broadcasted";
689 let summary = "Determines if 2 shapes can be successfully broadcasted";
/external/tensorflow/tensorflow/compiler/mlir/tensorflow/tests/
Dtf_optimize.mlir96 // The reshape splits broadcasted dimensions, instead of eliminating size-1 dimensions.
Dlower_tf.mlir434 // Check that inputs of different ranks are broadcasted and are lowered successfully.
/external/gemmlowp/doc/
Dpublic.md143 scalar. They are then broadcasted against LHS, RHS respectively.
/external/tensorflow/tensorflow/compiler/mlir/tensorflow/ir/
Dtf_device_ops.td207 packed_inputs: each input corresponds to an input broadcasted across all
Dtf_generated_ops.td1592 broadcasted shape. `s0`, `s1` and `r0` are all integer vectors.
1656 is broadcasted to output Tensor with shape of `[3, 3]`.
1658 When doing broadcasted operations such as multiplying a tensor
1660 benefit, as the broadcasted tensor is never materialized.
1663 The newly-created tensor takes the full memory of the broadcasted
3619 named axis label. If two inputs contain ellipsis, then they are broadcasted
3623 The broadcasted dimensions are placed in the corresponding location of the
3624 ellipsis in the output subscript. If the broadcasted dimensions are non-empty
17625 …F_Qint8, TF_Quint8, TF_Uint16, TF_Uint32, TF_Uint64, TF_Uint8]>, [{the broadcasted LHS tensor}]>:$…
17626 …F_Qint8, TF_Quint8, TF_Uint16, TF_Uint32, TF_Uint64, TF_Uint8]>, [{the broadcasted RHS tensor}]>:$…
/external/llvm-project/mlir/docs/
DTraits.md225 of the broadcasted operands. Specifically, starting from the most varying
/external/tensorflow/tensorflow/compiler/mlir/xla/transforms/
Dlegalize_tf_patterns.td85 // Check that two values can be broadcasted together
/external/autotest/docs/
Dwifi-basics-codelab.md434 being broadcasted to any receiver that can hear it. This is pattern is
/external/tensorflow/tensorflow/compiler/mlir/lite/ir/
Dtfl_ops.td4525 is broadcasted to output Tensor with shape of `[3, 3]`.
4527 When doing broadcasted operations such as multiplying a tensor
4529 benefit, as the broadcasted tensor is never materialized.
4532 The newly-created tensor takes the full memory of the broadcasted
/external/llvm-project/llvm/test/CodeGen/AMDGPU/
Dpacked-op-sel.ll30 ; Apply fneg to broadcasted vector
/external/llvm-project/llvm/docs/
DAMDGPUModifierSyntax.rst123 lane being broadcasted.
/external/tensorflow/tensorflow/compiler/mlir/lite/tests/
Dops.mlir1843 …@+1 {{'tfl.prelu' op result type '10x10' not broadcast compatible with broadcasted operands's shap…
1851 …1 {{'tfl.prelu' op result type '1x2x3x5' not broadcast compatible with broadcasted operands's shap…
/external/llvm-project/mlir/include/mlir/IR/
DOpBase.td2256 // Predicate to verify that the opId'th operand can be broadcasted to the type
/external/tensorflow/tensorflow/compiler/mlir/hlo/tests/
Dchlo_legalize_to_mhlo.mlir88 // Lower dynamically shaped `constant_like` to broadcasted constant.

12