- Sort Score
- Result 10 results
- Languages All
Results 31 - 40 of 50 for mat_mul (0.24 sec)
-
tensorflow/compiler/mlir/quantization/common/attrs_and_constraints_test.cc
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 14 17:10:32 UTC 2024 - 22.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/passes.h
bool enable_canonicalization, bool disable_fuse_mul_and_fc = false); std::unique_ptr<OperationPass<func::FuncOp>> CreateOptimizePass(); // Creates an instance of the Tensorflow Lite batch matmul Optimize pass. std::unique_ptr<OperationPass<func::FuncOp>> CreateOptimizeBatchMatmulPass(); // Creates an instance of the TensorFlow Lite dialect PrepareTF pass. std::unique_ptr<OperationPass<func::FuncOp>> CreatePrepareTFPass(
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Mar 07 21:29:34 UTC 2024 - 10.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/quantization/tensorflow/passes/prepare_lifting.cc
loc, value, Create1DConstValue(builder, loc, new_shape)); } return ConstantFoldOpIfPossible(value.getDefiningOp()).front(); } // Matches convolution op with "NHWC" data format or matmul op with false adj_y. // The list of supported ops in this function is: // - Conv2DOp // - Conv3DOp // - DepthwiseConv2dNativeOp // - MatMulOp // - BatchMatMulV2Op
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri May 17 17:58:54 UTC 2024 - 13.3K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/passes.td
"bool", "false", "Disable folding mul and fully connected ops during optimization pass.">, ]; } def OptimizeBatchMatmulPass : Pass<"tfl-optimize-batch-matmul", "mlir::func::FuncOp"> { let summary = "Optimize FC with BatchMatmul within the TensorFlow Lite dialect"; let constructor = "CreateOptimizeBatchMatmulPass()"; let dependentDialects = ["TFL::TensorFlowLiteDialect"]; }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Apr 24 20:30:06 UTC 2024 - 22.6K bytes - Viewed (0) -
tensorflow/compiler/mlir/tfrt/tests/analysis/cost_analysis.mlir
// 262657 = 1 + 512 + 512 * 512 // expected-remark@+1 {{Cost: 262657}} %2 = "tf.MatMul"(%arg, %1) {device = "/job:localhost/replica:0/task:0/device:CPU:0", transpose_a = false, transpose_b = false} : (tensor<?x512xf32>, tensor<512x512xf32>) -> tensor<?x512xf32> // expected-remark@+1 {{Cost: 512}}
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Aug 14 15:35:49 UTC 2023 - 12.2K bytes - Viewed (0) -
tensorflow/compiler/mlir/quantization/tensorflow/passes/passes.h
std::unique_ptr<OperationPass<mlir::func::FuncOp>> CreateOptimizePass(); // Creates an instance of the ReplaceCastHacksWithTFXLAOpsPass, which will // replace mixed-type convolution and matmul cast hacks by XLA Conv2DOp and // MatmulOp. std::unique_ptr<OperationPass<func::FuncOp>> CreateReplaceCastHacksWithTFXLAOpsPass(); // Creates a pass that moves & merges initializer function's ops into the @main
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri May 10 04:07:09 UTC 2024 - 12.3K bytes - Viewed (0) -
src/crypto/aes/gcm_amd64.s
JMP dataOctaLoop dataTLS: MOVOU (16*14)(pTbl), T1 MOVOU (16*15)(pTbl), T2 PXOR B0, B0 MOVQ (aut), B0 PINSRD $2, 8(aut), B0 PINSRB $12, 12(aut), B0 XORQ autLen, autLen JMP dataMul dataOctaLoop: CMPQ autLen, $128 JB startSinglesLoop SUBQ $128, autLen MOVOU (16*0)(aut), X0 MOVOU (16*1)(aut), X1 MOVOU (16*2)(aut), X2 MOVOU (16*3)(aut), X3
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Mon Mar 04 17:29:44 UTC 2024 - 23.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/legalize_tf.cc
DECL_CONVERT_OP(Assert); DECL_CONVERT_OP(ConcatV2); DECL_CONVERT_OP(BatchMatMul); DECL_CONVERT_OP(BatchMatMulV2); DECL_CONVERT_OP(BatchMatMulV3); DECL_CONVERT_OP(MatMul); DECL_CONVERT_OP(MatrixDiagV2); DECL_CONVERT_OP(MatrixDiagV3); DECL_CONVERT_OP(Pack); DECL_CONVERT_OP(Split); DECL_CONVERT_OP(SplitV); DECL_CONVERT_OP(Unpack); DECL_CONVERT_OP(Conv3D);
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon May 20 20:06:54 UTC 2024 - 45.2K bytes - Viewed (0) -
src/crypto/aes/gcm_arm64.s
VLD1.P (pTbl), [T1.B16, T2.B16] VEOR B0.B16, B0.B16, B0.B16 MOVD (aut), H0 VMOV H0, B0.D[0] MOVW 8(aut), H0 VMOV H0, B0.S[2] MOVB 12(aut), H0 VMOV H0, B0.B[12] MOVD $0, autLen B dataMul octetsLoop: CMP $128, autLen BLT startSinglesLoop SUB $128, autLen VLD1.P 32(aut), [B0.B16, B1.B16] VLD1.P 32(pTbl), [T1.B16, T2.B16] VREV64 B0.B16, B0.B16
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Mon Mar 04 17:29:44 UTC 2024 - 21.5K bytes - Viewed (0) -
tensorflow/compiler/mlir/tfrt/tests/tf_to_corert/control_flow.mlir
%x = "tf.TensorArrayReadV3"(%handle_0, %index, %flow_0) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<2x!tf_type.resource<tensor<?x100xf32>>>, tensor<i32>, tensor<f32>) -> tensor<?x100xf32> %y = "tf.MatMul"(%x, %cst) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<?x100xf32>, tensor<100x512xf32>) -> (tensor<?x512xf32>)
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 14 00:40:32 UTC 2024 - 17.5K bytes - Viewed (0)