Tensor (full) in projects: aosp_15_r20 - OpenGrok search results

Project(s)

Full Search
Definition
Symbol
File Path
History
Type

Searched full:tensor (Results 1 – 25 of 12798) sorted by relevance

12 3 4 5 6 7 8 9 10 >>...512

/aosp_15_r20/external/tensorflow/tensorflow/core/transforms/graph_compactor/tests/
H A D	rename_lots.mlir	`12 %arg0: tensor<i1>, 13 %arg1: tensor<i1>, 14 %arg2: tensor<i1>, 15 %arg3: tensor<i1>, 16 %arg4: tensor<i1>, 17 %arg5: tensor<i1>, 18 %arg6: tensor<i1>, 19 %arg7: tensor<i1>, 20 %arg8: tensor<i1>, 21 %arg9: tensor<i1>, [all …]`
/aosp_15_r20/external/pytorch/torch/csrc/autograd/
H A D	FunctionsManual.h	34 TORCH_API Tensor toNonOptFwGrad(const std::optional<Tensor>& t); 35 TORCH_API Tensor toNonOptPrimal(const std::optional<Tensor>& t); 36 TORCH_API Tensor toNonOptTensor(const std::optional<Tensor>& t); 38 TORCH_API inline std::optional<Tensor> wrap_opt_if( in wrap_opt_if() 39 const Tensor& t, in wrap_opt_if() 41 using OptTensor = std::optional<Tensor>; in wrap_opt_if() 45 TORCH_API Tensor 46 apply_loss_reduction(const Tensor& unreduced, int64_t reduction); 51 const at::Tensor& t); 55 at::ArrayRef<at::Tensor> t); [all …]
/aosp_15_r20/external/pytorch/docs/source/
H A D	tensors.rst	5 torch.Tensor 8 A :class:`torch.Tensor` is a multi-dimensional matrix containing elements of 15 Torch defines tensor types with the following data types: 69 Data type CPU tensor GPU tensor 85 :class:`torch.Tensor` constructor is an alias for the default tensor type 91 A tensor can be constructed from a Python :class:`list` or sequence using the 92 :func:`torch.tensor` constructor: 96 >>> torch.tensor([[1., -1.], [1., -1.]]) 97 tensor([[ 1.0000, -1.0000], 99 >>> torch.tensor(np.array([[1, 2, 3], [4, 5, 6]])) [all …]
/aosp_15_r20/external/pytorch/aten/src/ATen/native/
H A D	native_functions.yaml	9 - func: _cast_Byte(Tensor self, bool non_blocking=False) -> Tensor 13 - func: _cast_Char(Tensor self, bool non_blocking=False) -> Tensor 17 - func: _cast_Double(Tensor self, bool non_blocking=False) -> Tensor 21 - func: _cast_Float(Tensor self, bool non_blocking=False) -> Tensor 25 - func: _cast_Int(Tensor self, bool non_blocking=False) -> Tensor 29 - func: _cast_Long(Tensor self, bool non_blocking=False) -> Tensor 33 - func: _cast_Short(Tensor self, bool non_blocking=False) -> Tensor 37 - func: _cast_Half(Tensor self, bool non_blocking=False) -> Tensor 40 # Computes the gradient of current tensor w.r.t. graph leaves. 41 - func: _backward(Tensor self, Tensor[] inputs, Tensor? gradient=None, bool? retain_graph=None, boo… [all …]
/aosp_15_r20/external/tensorflow/tensorflow/compiler/mlir/lite/tests/
H A D	fuse-tftext.mlir	3 …ce_tokenizer_rank1(%arg0: tensor<1x!tf_type.string> {tf._user_specified_name = "input"}) -> (tenso… 4 %0 = "tf.Const"() {value = dense<[0, 1]> : tensor<2xi64>} : () -> tensor<2xi64> 5 %1 = "tf.Const"() {value = dense<[]> : tensor<0xi64>} : () -> tensor<0xi64> 6 %2 = "tf.Const"() {value = dense<true> : tensor<i1>} : () -> tensor<i1> 7 %3 = "tf.Const"() {value = dense<-1> : tensor<i32>} : () -> tensor<i32> 8 %4 = "tf.Const"() {value = dense<[[0], [1]]> : tensor<2x1xi64>} : () -> tensor<2x1xi64> 9 %5 = "tf.Const"() {value = dense<-1> : tensor<1xi32>} : () -> tensor<1xi32> 10 %6 = "tf.Const"() {value = dense<2> : tensor<1xi32>} : () -> tensor<1xi32> 11 %7 = "tf.Const"() {value = dense<1> : tensor<i32>} : () -> tensor<i32> 12 %8 = "tf.Const"() {value = dense<2> : tensor<i32>} : () -> tensor<i32> [all …]
H A D	legalize-tf.mlir	3 func.func @add(%arg0: tensor<1xf32>, %arg1: tensor<1xf32>) -> tensor<1xf32> { 4 %0 = "tf.Add"(%arg0, %arg1) : (tensor<1xf32>, tensor<1xf32>) -> tensor<1xf32> 5 func.return %0: tensor<1xf32> 8 // CHECK: tfl.add %arg0, %arg1 {fused_activation_function = "NONE"} : tensor<1xf32> 12 func.func @sub(%arg0: tensor<1xi64>, %arg1: tensor<1xi64>) -> tensor<1xi64> { 13 %0 = "tf.Sub"(%arg0, %arg1) : (tensor<1xi64>, tensor<1xi64>) -> tensor<1xi64> 14 func.return %0: tensor<1xi64> 17 // CHECK: tfl.sub %arg0, %arg1 {fused_activation_function = "NONE"} : tensor<1xi64> 22 …c @testAddHighDimsHaveSameShape(%arg0: tensor<1x2x3x4x5x6x7x8xi32>, %arg1: tensor<1x2x3x4x5x6x7x8x… 24 …%0 = "tf.Add"(%arg0, %arg1) : (tensor<1x2x3x4x5x6x7x8xi32>, tensor<1x2x3x4x5x6x7x8xi32>) -> tensor… [all …]
H A D	ops.mlir	7 func.func @testCos(tensor<? x f32>) -> tensor<? x f32> { 8 ^bb0(%arg0: tensor<? x f32>): 10 %0 = "tfl.cos"(%arg0): (tensor<? x f32>) -> tensor<? x f32> 11 func.return %0 : tensor<? x f32> 17 func.func @testCosWithWrongInputType(tensor<?xi32>) -> tensor<?xi32> { 18 ^bb0(%arg0: tensor<?xi32>): 19 // expected-error @+1 {{tfl.cos' op operand #0 must be tensor of 32-bit float values}} 20 %0 = "tfl.cos"(%arg0): (tensor<?xi32>) -> tensor<?xi32> 21 func.return %0#0 : tensor<?xi32> 27 func.func @testExp(tensor<? x f32>) -> tensor<? x f32> { [all …]
H A D	optimize.mlir	10 …unc.func @fusedConv2dRelu(%arg0: tensor<256x32x32x3xf32>, %arg1: tensor<16x3x3x3xf32>, %arg2: tens… 11 …ide_h = 1 : i32, stride_w = 1 : i32} : (tensor<256x32x32x3xf32>, tensor<16x3x3x3xf32>, tensor<16xf… 12 %1 = "tfl.relu"(%0) : (tensor<256x32x32x16xf32>) -> tensor<256x32x32x16xf32> 13 func.return %1 : tensor<256x32x32x16xf32> 15 …ide_h = 1 : i32, stride_w = 1 : i32} : (tensor<256x32x32x3xf32>, tensor<16x3x3x3xf32>, tensor<16xf… 20 …fusedDepthwiseConv2dRelu6(%arg0: tensor<256x32x32x3xf32>, %arg1: tensor<16x3x3x3xf32>, %arg2: tens… 21 …ide_h = 4 : i32, stride_w = 5 : i32} : (tensor<256x32x32x3xf32>, tensor<16x3x3x3xf32>, tensor<16xf… 22 %1 = "tfl.relu6"(%0) : (tensor<256x30x30x16xf32>) -> tensor<256x30x30x16xf32> 23 func.return %1 : tensor<256x30x30x16xf32> 25 …ide_h = 4 : i32, stride_w = 5 : i32} : (tensor<256x32x32x3xf32>, tensor<16x3x3x3xf32>, tensor<16xf… [all …]
H A D	lower-static-tensor-list.mlir	1 // RUN: tf-opt "-tfl-lower-static-tensor-list=allow-tensorlist-pass-through default-to-single-batch… 6 func.func @tensorlistConst(%arg0 : tensor<1xi32>) -> tensor<2x3xi32> { 7 …G: %[[ELEMENT0:.]] = "tf.Const"() {value = dense<[0, 1, 2]> : tensor<3xi32>} : () -> tensor<3xi32> 8 …G: %[[ELEMENT1:.]] = "tf.Const"() {value = dense<[3, 4, 5]> : tensor<3xi32>} : () -> tensor<3xi32> 9 ….Pack"(%[[ELEMENT0]], %[[ELEMENT1]]) {axis = 0 : i64} : (tensor<3xi32>, tensor<3xi32>) -> tensor<2… 10 …0333A5C3030335C3030335C3030345C30303522"> : tensor<!tf_type.variant>} : () -> tensor<!tf_type.vari… 13 …%1 = "tf.TensorListStack"(%0, %arg0) : (tensor<!tf_type.variant<tensor<3xi32>>>, tensor<1xi32>) ->… 14 func.return %1 : tensor<2x3xi32> 20 func.func @emptyTensorlistConst(%arg0 : tensor<1xi32>) -> tensor<0x3xi32> { 21 …030315C3032325C3030325C3031305C30303322"> : tensor<!tf_type.variant>} : () -> tensor<!tf_type.vari… [all …]
H A D	prepare-composite-functions-tf.mlir	4 func.func @embedding(%arg0: tensor<xf32>, %arg1: tensor<xi32>) -> tensor<xf32> attributes {tf._… 5 %0 = "tf.Const"() {value = dense<1> : tensor<i32>} : () -> tensor<i32> 6 %1 = "tf.ExpandDims"(%arg1, %0) : (tensor<xi32>, tensor<i32>) -> tensor<xi32> 7 %2 = "tf.Const"() {value = dense<1> : tensor<i32>} : () -> tensor<i32> 8 %3 = "tf.Const"() {value = dense<4096> : tensor<i32>} : () -> tensor<i32> 9 %4 = "tf.Const"() {value = dense<0> : tensor<i32>} : () -> tensor<i32> 10 %5 = "tf.Range"(%4, %3, %2) : (tensor<i32>, tensor<i32>, tensor<i32>) -> tensor<4096xi32> 11 %6 = "tf.Equal"(%1, %5) : (tensor<xi32>, tensor<4096xi32>) -> tensor<xi1> 12 %7 = "tf.Cast"(%6) : (tensor<xi1>) -> tensor<xf32> 13 …chMatMulV2"(%7, %arg0) {adj_x = false, adj_y = false} : (tensor<xf32>, tensor<xf32>) -> tensor<… [all …]
H A D	dilated-conv.mlir	3 func.func @testDilatedConv(%arg0: tensor<1x128x128x3xf32>, %arg1: tensor<5x5x3x8xf32>) -> tensor<1x… 4 %cst = arith.constant dense<[2, 2]> : tensor<2xi32> 5 %cst_0 = arith.constant dense<4> : tensor<2x2xi32> 6 …tf.SpaceToBatchND"(%arg0, %cst, %cst_0) : (tensor<1x128x128x3xf32>, tensor<2xi32>, tensor<2x2xi32>… 7 …1) {padding = "VALID", strides = [1, 1, 1, 1]} : (tensor<4x68x68x3xf32>, tensor<5x5x3x8xf32>) -> t… 8 … = "tf.BatchToSpaceND"(%1, %cst, %cst_0) : (tensor<4x64x64x8xf32>, tensor<2xi32>, tensor<2x2xi32>)… 9 func.return %2 : tensor<1x120x120x8xf32> 12 // CHECK-SAME: ([[INPUT:%.]]: tensor<1x128x128x3xf32>, [[FILTER:%.]]: tensor<5x5x3x8xf32>) 13 …], padding = "VALID", strides = [1, 1, 1, 1]} : (tensor<1x128x128x3xf32>, tensor<5x5x3x8xf32>) -> … 14 // CHECK-NEXT: return [[RESULT]] : tensor<1x120x120x8xf32> [all …]
/aosp_15_r20/external/tensorflow/tensorflow/compiler/mlir/tensorflow/tests/
H A D	shape_inference.mlir	4 // CHECK-LABEL: func @main(%arg0: tensor<1xi32>, %arg1: tensor<1xi32>) -> tensor<1xi32> 5 func.func @main(%arg0: tensor<1xi32>, %arg1: tensor<1xi32>) -> tensor<xi32> { 7 // CHECK-SAME: (tensor<1xi32>, tensor<1xi32>) -> tensor<1xi32> 8 // CHECK: return %[[RESULT]] : tensor<1xi32> 9 %0 = "tf.Cast"(%arg0) : (tensor<1xi32>) -> tensor<xi32> 10 %1 = "tf.Cast"(%arg1) : (tensor<1xi32>) -> tensor<xi32> 11 %2 = "tf.AddV2"(%0, %1) : (tensor<xi32>, tensor<xi32>) -> tensor<xi32> 12 func.return %2 : tensor<xi32> 16 func.func @simple_chain(%arg0: tensor<1xf32>) -> tensor<xf32> { 17 // CHECK: %[[MUL:.]] = "tf.Mul"{{.}} (tensor<1xf32>, tensor<1xf32>) -> tensor<1xf32> [all …]
H A D	tf-ops.mlir	17 // CHECK: "tf.TensorProtoIntTensor"() {bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x1x4… 18 …"tf.TensorProtoIntTensor"(){bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x1x4xi32>} : (… 19 // CHECK: "tf.TensorProtoFloatTensor"() {bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x1… 20 …"tf.TensorProtoFloatTensor"(){bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x1x4xf32>} :… 21 // CHECK: "tf.TensorProtoStringTensor"() {bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x… 22 …"tf.TensorProtoStringTensor"(){bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x1x4x!tf_ty… 23 // CHECK: "tf.TensorProtoResourceTensor"() {bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<… 24 …"tf.TensorProtoResourceTensor"(){bar = #tf_type<tensor_proto : "0x68656C6C6F"> : tensor<2x1x4x!tf_… 45 func.func @testIdentity(%arg0: tensor<4x?x!tf_type.stringref>) -> tensor<4x2x!tf_type.string> { 46 %0 = "tf.Identity"(%arg0) : (tensor<4x?x!tf_type.stringref>) -> tensor<4x2x!tf_type.string> [all …]
H A D	canonicalize.mlir	4 func.func @tfAssertTrue(%arg0: tensor<1x1x6x2xf32>) { 5 %t = arith.constant dense<true> : tensor<i1> 7 "tf.Assert"(%t, %arg0) {summarize = 3} : (tensor<i1>, tensor<1x1x6x2xf32>) -> () 12 func.func @tfAssertFalse(%arg0: tensor<1x1x6x2xf32>) { 13 %f = arith.constant dense<false> : tensor<i1> 15 "tf.Assert"(%f, %arg0) {summarize = 3} : (tensor<i1>, tensor<1x1x6x2xf32>) -> () 21 func.func @testGatherToV2(%params: tensor<4x3xf32>, %indices: tensor<1x2xi32>) -> tensor<2x3xf32> { 22 // CHECK: %[[AXIS:.*]] = "tf.Const"() {value = dense<0> : tensor<i32>} : () -> tensor<i32> 23 …g0, %arg1, %[[AXIS]]) {batch_dims = 0 : i64} : (tensor<4x3xf32>, tensor<1x2xi32>, tensor<i32>) -> … 24 %0 = "tf.Gather"(%params, %indices) : (tensor<4x3xf32>, tensor<1x2xi32>) -> tensor<2x3xf32> [all …]
H A D	merge_control_flow.mlir	11 %0 = "tf.Const"() {value = dense<true> : tensor<i1>} : () -> tensor<i1> 12 %1 = "tf.Const"() {value = dense<false> : tensor<i1>} : () -> tensor<i1> 14 %2 = "tf.A"() : () -> (tensor<f32>) 18 }) {is_stateless = true} : (tensor<i1>) -> () 20 %2 = "tf.B"() : () -> (tensor<f32>) 24 }) {is_stateless = true} : (tensor<i1>) -> () 38 %0 = "tf.Const"() {value = dense<true> : tensor<i1>} : () -> tensor<i1> 39 %1 = "tf.Const"() {value = dense<false> : tensor<i1>} : () -> tensor<i1> 40 %3 = "tf.A"() : () -> (tensor<?xf32>) 41 %4 = "tf.B"() : () -> (tensor<i32>) [all …]
H A D	legalize_hlo.mlir	6 // CHECK-SAME: %[[VAL_0:.]]: tensor<1x32x10x32xi32>, 7 // CHECK-SAME: %[[VAL_1:.]]: tensor<32xi32>) -> tensor<1x32x10x32xi32> { 8 …[VAL_2:.]] = "tf.AddV2"(%[[VAL_0]], %[[VAL_1]]) : (tensor<1x32x10x32xi32>, tensor<32xi32>) -> ten… 9 // CHECK: return %[[VAL_2]] : tensor<1x32x10x32xi32> 11 func.func @biasAdd_NHWC(%arg0: tensor<1x32x10x32xi32>, %arg1: tensor<32xi32>) -> tensor<1x32x10x32x… 12 … %arg1) {broadcast_dimensions = dense<3> : tensor<1xi64>} : (tensor<1x32x10x32xi32>, tensor<32xi32… 13 func.return %0 : tensor<1x32x10x32xi32> 17 // CHECK-SAME: %[[VAL_0:.]]: tensor<1x32x10x32xi32>, 18 // CHECK-SAME: %[[VAL_1:.]]: tensor<32xi32>) -> tensor<1x32x10x32xi32> { 19 …[VAL_2:.]] = "tf.AddV2"(%[[VAL_0]], %[[VAL_1]]) : (tensor<1x32x10x32xi32>, tensor<32xi32>) -> ten… [all …]
/aosp_15_r20/external/tensorflow/tensorflow/compiler/xla/mlir_hlo/stablehlo/tests/
H A D	verify_reduce_window.mlir	4 func.func @reduce_window(%arg0: tensor<4x2xf32>, %arg1: tensor<4x2xi32>, 5 %init0: tensor<f32>, %init1: tensor<i32>) -> 6 (tensor<2x2xf32>, tensor<2x2xi32>) { 8 ^bb0(%a0: tensor<f32>, %a1: tensor<i32>, 9 %b0: tensor<f32>, %b1: tensor<i32>): 10 %2 = stablehlo.add %a0, %b0 : tensor<f32> 11 %3 = stablehlo.add %a1, %b1 : tensor<i32> 12 "stablehlo.return"(%2, %3) : (tensor<f32>, tensor<i32>) -> () 14 { padding = dense<[[2, 2], [0, 0]]> : tensor<2x2xi64>, 15 window_dimensions = dense<[5, 1]> : tensor<2xi64>, [all …]
H A D	ops_stablehlo.mlir	17 func.func @reduce_scatter(%data: tensor<4x16xf32>) -> tensor<4x4xf32> { 20 ^bb0(%arg2: tensor<f32>, %arg3: tensor<f32>): 21 %1 = stablehlo.add %arg2, %arg3 : tensor<f32> 22 "stablehlo.return"(%1) : (tensor<f32>) -> () 23 }) {replica_groups = dense<[[0, 1, 2, 3]]> : tensor<1x4xi64>, 24 scatter_dimension = 1 : i64} : (tensor<4x16xf32>) -> tensor<4x4xf32> 25 func.return %0 : tensor<4x4xf32> 30 func.func @invalid_reduce_scatter(%data: tensor<4x16xf32>) -> tensor<4x5xf32> { 34 ^bb0(%arg2: tensor<f32>, %arg3: tensor<f32>): 35 %1 = stablehlo.add %arg2, %arg3 : tensor<f32> [all …]
/aosp_15_r20/external/tensorflow/tensorflow/compiler/xla/mlir_hlo/tests/Dialect/mhlo/
H A D	ops.mlir	17 func.func @reduce_scatter(%data: tensor<4x16xf32>) -> tensor<4x4xf32> { 20 ^bb0(%arg2: tensor<f32>, %arg3: tensor<f32>): 21 %1 = mhlo.add %arg2, %arg3 : tensor<f32> 22 "mhlo.return"(%1) : (tensor<f32>) -> () 23 }) {replica_groups = dense<[[0, 1, 2, 3]]> : tensor<1x4xi64>, 24 scatter_dimension = 1 : i64} : (tensor<4x16xf32>) -> tensor<4x4xf32> 25 func.return %0 : tensor<4x4xf32> 30 func.func @invalid_reduce_scatter(%data: tensor<4x16xf32>) -> tensor<4x5xf32> { 34 ^bb0(%arg2: tensor<f32>, %arg3: tensor<f32>): 35 %1 = mhlo.add %arg2, %arg3 : tensor<f32> [all …]
H A D	verifier_reduce_window_op.mlir	4 func.func @reduce_window(%arg0: tensor<4x2xf32>, %arg1: tensor<4x2xi32>, 5 %init0: tensor<f32>, %init1: tensor<i32>) -> 6 (tensor<2x2xf32>, tensor<2x2xi32>) { 8 ^bb0(%a0: tensor<f32>, %a1: tensor<i32>, 9 %b0: tensor<f32>, %b1: tensor<i32>): 10 %2 = mhlo.add %a0, %b0 : tensor<f32> 11 %3 = mhlo.add %a1, %b1 : tensor<i32> 12 "mhlo.return"(%2, %3) : (tensor<f32>, tensor<i32>) -> () 14 { padding = dense<[[2, 2], [0, 0]]> : tensor<2x2xi64>, 15 window_dimensions = dense<[5, 1]> : tensor<2xi64>, [all …]
H A D	verifier_reduce_op.mlir	8 func.func @reduce_valid(%arg0: tensor<4x4xf32>, %arg1 : tensor<4xf32>) 9 -> (tensor<4xf32>) { 12 ^bb0(%arg2: tensor<4xf32>, %arg3: tensor<4xf32> ): 13 %1 = "mhlo.add"(%arg2, %arg3) : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32> 14 "mhlo.return"(%1) : (tensor<4xf32>) -> () 16 }) {dimensions = dense<[0]> : tensor<1xi64>} : (tensor<4x4xf32>, tensor<4xf32>) -> tensor<4xf32> 18 func.return %0: tensor<4xf32> 24 func.func @reduce_complex_type(%arg0: tensor<1x2xcomplex<f32>>, %arg1 : tensor<complex<f32>>) 25 -> (tensor<1xcomplex<f32>>) { 28 ^bb0(%arg2: tensor<complex<f32>> loc("foo"), %arg3: tensor<complex<f32>> loc("foo")): [all …]
/aosp_15_r20/external/tensorflow/tensorflow/compiler/mlir/hlo/tests/
H A D	mhlo_flatten_tuple.mlir	4 // CHECK-SAME: %arg0: tensor<3xf32>) -> tensor<3xf32> { 5 // CHECK: %[[CST_0:.]] = constant dense<0> : tensor<1xi32> 6 // CHECK: %[[CST_1:.]] = constant dense<100> : tensor<2xi32> 7 // CHECK: %[[CST_2:.*]] = constant dense<1.000000e+00> : tensor<1xf32> 9 // CHECK: ^bb0(%arg1: tensor<1xi32>, %arg2: tensor<2xi32>, %arg3: tensor<1xf32>, %arg4: te… 10 …es = dense<1> : tensor<1xi64>, start_indices = dense<0> : tensor<1xi64>, strides = dense<1> : tens… 11 …e"(%arg1, %[[SLICE_0]]) {comparison_direction = "LT"} : (tensor<1xi32>, tensor<1xi32>) -> tensor<1… 12 // CHECK: "mhlo.return"(%[[COMPARE_0]]) : (tensor<1xi1>) -> () 14 // CHECK: ^bb0(%arg1: tensor<1xi32>, %arg2: tensor<2xi32>, %arg3: tensor<1xf32>, %arg4: te… 15 …cast_in_dim"(%arg3) {broadcast_dimensions = dense<0> : tensor<1xi64>} : (tensor<1xf32>) -> tensor<… [all …]
/aosp_15_r20/external/pytorch/tools/autograd/
H A D	derivatives.yaml	32 # should contain only booleans, specifying whether each of the output Tensor 40 # There are two cases for Tensor and TensorList arguments here: 87 # - Any of the input arguments, tensor or non-tensor, including 103 # specifying if either zero or at least one tensor from the list requires 113 # - `wrap_opt_if`, is a 2-argument function that accepts a tensor 115 # variable in a graph. The result of this function is `c10::optional<Tensor>`, 117 # otherwise it is the variable wrapped in `c10::optional<Tensor>`. 157 # to check if any computation is needed and should return an undefined Tensor when 168 # - If a function return at least one Tensor that is a differentiable view 170 # - If there is only one differentiable output, this Tensor is marked as a [all …]
/aosp_15_r20/external/tensorflow/tensorflow/compiler/xla/mlir_hlo/tests/Dialect/mhlo/canonicalize/
H A D	canonicalize.mlir	4 func.func @add_fold() -> tensor<4xi64> { 5 %0 = mhlo.constant dense<[1, 2, 3, 4]> : tensor<4xi64> 6 %1 = mhlo.constant dense<[5, 6, 7, 8]> : tensor<4xi64> 8 %2 = "mhlo.add"(%0, %1) : (tensor<4xi64>, tensor<4xi64>) -> (tensor<4xi64>) 9 func.return %2 : tensor<4xi64> 13 func.func @add_scalar_fold() -> tensor<4xi64> { 14 %0 = mhlo.constant dense<1> : tensor<4xi64> 15 %1 = mhlo.constant dense<5> : tensor<4xi64> 17 %2 = "mhlo.add"(%0, %1) : (tensor<4xi64>, tensor<4xi64>) -> (tensor<4xi64>) 18 func.return %2 : tensor<4xi64> [all …]
/aosp_15_r20/external/executorch/exir/dialects/edge/op/
H A D	sample_input.py	16 "_log_softmax.default": { # (Tensor self, int dim, bool half_to_float) -> Tensor 18 InArg(ArgType.Tensor), 23 Return(ArgType.Tensor), 26 … { # (Tensor input, Tensor? weight, Tensor? bias, Tensor running_mean, Tensor running_var, float … 28 InArg(ArgType.Tensor, size=[2, 3, 4, 5]), 31 InArg(ArgType.Tensor, size=[3]), 32 InArg(ArgType.Tensor, size=[3]), 37 Return(ArgType.Tensor, argname="__ret0", size=[2, 3, 4, 5]), 38 Return(ArgType.Tensor, argname="__ret1", size=[0]), 39 Return(ArgType.Tensor, argname="__ret2", size=[0]), [all …]

12 3 4 5 6 7 8 9 10 >>...512