- Sort Score
- Result 10 results
- Languages All
Results 1 - 10 of 2,961 for FusedN (0.18 sec)
-
tensorflow/compiler/mlir/tfr/tests/decompose.mlir
// expected-error@+1 {{Denied unregistered attribute was found: denied_attr}} %0:2 = "tf.FusedN"(%arg0, %arg1, %arg2) {A=0:index, denied_attr} : (tensor<1x2x3x4x!tf_type.string>, tensor<f32>, tensor<f32>) -> (tensor<1x2x3x4x!tf_type.string>, tensor<f32>) func.return %0#1 : tensor<f32> // CHECK-NEXT: "tf.FusedN"(%arg0, %arg1, %arg2) {A = 0 : index, denied_attr} } // CHECK-LABEL: quantized_tensor
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Oct 30 06:52:55 UTC 2023 - 16.7K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/transforms/fused_kernel_matcher.cc
contraction, "fused operation must be nested inside a function, If or While"); } // If the contraction is used in multiple places, fusing it will only create // more contraction nodes, which is slower. if (!contraction.getResult().hasOneUse()) return rewriter.notifyMatchFailure(contraction, "result is used by multiple ops");
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 14.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/quantization/common/quantization_lib/quantization_config.h
// DT_FLOAT, DT_HALF, DT_QINT8, and DT_QUINT8. When DT_HALF is used, the // `weight_quantization` flag needs to set to true. When DT_QUINT8 is used, // the `weight_quantization` flag needs to set to false. tensorflow::DataType inference_type = tensorflow::DT_FLOAT; // The input and output data type during inference. This flag is only used // when `inference_type` is different from DT_FLOAT. This flag can only be set
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Mar 13 10:16:19 UTC 2024 - 10.8K bytes - Viewed (0) -
src/vendor/golang.org/x/sys/cpu/cpu.go
HasAVX512IFMA bool // Advanced vector extension 512 Integer Fused Multiply Add HasAVX512VBMI bool // Advanced vector extension 512 Vector Byte Manipulation Instructions HasAVX5124VNNIW bool // Advanced vector extension 512 Vector Neural Network Instructions Word variable precision HasAVX5124FMAPS bool // Advanced vector extension 512 Fused Multiply Accumulation Packed Single precision
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Wed May 08 16:12:58 UTC 2024 - 12.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/prepare_composite_functions_tf.cc
// TFLite fused embedding_lookup op. ConvertEmbeddedLookupFunc convert_embedded_lookup(func); if (failed(convert_embedded_lookup.VerifySignature())) return; func.eraseBody(); func.addEntryBlock(); convert_embedded_lookup.RewriteFunc(); } else if (attr.getValue() == mlir::TFL::kLstmCellSimple) { // Check if the lstm cell simple can be fused, if not, we just don't do // anything.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 17.6K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/passes.td
"comma separated list of allowlisted functions to be quantized. Only used in tests">, Option<"quantize_signed_", "quantize-signed", "bool", "false", "signed inference type. Only used in tests">, Option<"activation_number_of_bits_", "activation-number-of-bits", "int", "8", "number of bits for inference type. Only used in tests">,
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Apr 24 20:30:06 UTC 2024 - 22.6K bytes - Viewed (0) -
platforms/core-runtime/BYTECODE-INTERCEPTION-README.md
metadata("1. Artifact metadata: hash, name\n2. Dependencies instrumented type hierarchy") result("1. Transformed jar\n2. hash of original jar") buildService1[["Build service 1:\nused to provide type hierarchies of \ndependencies"]] buildService2[["Build service 2:\nused to get original Jar\nvia hash from original classpath"]] jar --> collect --> artifactOutput --> merge --> metadata --> instrument --> result
Registered: Wed Jun 12 18:38:38 UTC 2024 - Last Modified: Thu Apr 18 09:22:58 UTC 2024 - 22.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/quantization/tensorflow/tests/unfreeze_constants.mlir
// CHECK: return %[[CONCAT]] : tensor<12xf32> } // ----- // Tests a case where the ConstOp's location is a fused loc containing more // than two strings to be combined to form the shared_name. It must not contain // the character ";" (which is often used as a delimiter to join fused loc's // items). module attributes {tf_saved_model.semantics} { // CHECK: func.func @init_func_restore_op()
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Oct 30 06:52:55 UTC 2023 - 17.2K bytes - Viewed (0) -
tensorflow/compiler/mlir/quantization/stablehlo/passes/passes.td
"mlir::stablehlo::StablehloDialect", ]; } def UnwrapXlaCallModuleOpPass : Pass<"stablehlo-unwrap-xla-call-module-op", "ModuleOp"> { let summary = "Unwrap XlaCallModuleOps into inline functions if not used for quantizing fused patterns."; let dependentDialects = ["TF::TensorFlowDialect"]; } def ConvertFuncToBfloat16Pass : Pass<"stablehlo-convert-func-to-bfloat16", "mlir::func::FuncOp"> {
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 14 06:31:57 UTC 2024 - 10.3K bytes - Viewed (0) -
src/cmd/compile/internal/test/testdata/fp_test.go
{0.29711226, 0.89436173, 0.097454615, 0.36318043}, // fused multiply-add result: 0.36318046 {0.6810783, 0.24151509, 0.31152245, 0.47601312}, // fused multiply-add result: 0.47601315 {0.73023146, 0.18292491, 0.4283571, 0.5619346}, // fused multiply-add result: 0.56193465 {0.89634174, 0.32208398, 0.7211478, 1.009845}, // fused multiply-add result: 1.0098451
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Tue Apr 05 17:54:15 UTC 2022 - 35K bytes - Viewed (0)