- Sort Score
- Result 10 results
- Languages All
Results 91 - 100 of 247 for legalize (0.12 sec)
-
tensorflow/compiler/mlir/lite/transforms/optimize_patterns.td
(HasOneUse $relu_out), (HasOneUse $mul_out), (HasOneUse $input_neg_out)]>; // The constant folding in this pass might produce constant in the tf dialect. // This rule is to legalize these constant to the tfl dialect. def LegalizeConstOp : Pat< (TF_ConstOp ElementsAttr:$value), (TFL_ConstOp $value)>; // Reorders adds to allow constant folding. // Add --> Add $input, $constantA
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu May 16 20:31:41 UTC 2024 - 66.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/transforms/xla_legalize_tf.cc
"/tensorflow/core/tf2xla/v1/mlir_failed_xla_legalize_tf_pass_count", "Counts the failure of legalization of ops", "op_name", "legality"); class LegalizeTF : public impl::LegalizeTFBase<LegalizeTF> { public: explicit LegalizeTF(bool legalize_chlo, std::optional<StringRef> tf2xla_fallback_device_type, bool prefer_tf2xla) { legalize_chlo_ = legalize_chlo;
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 28 21:49:50 UTC 2024 - 10.2K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/tests/optimize.mlir
// Run legalize pass and then optimize pass, and make sure some fusing is applied. // RUN: tf-opt %s -tfl-legalize-tf -tfl-optimize | FileCheck --check-prefix=Fusing %s // Run legalize pass and then optimize pass, and make sure some fusing is applied, but no mul->fc. // RUN: tf-opt %s -tfl-legalize-tf -tfl-optimize='disable-fuse-mul-and-fc=true' | FileCheck --check-prefix=NoFusing %s
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu May 16 20:31:41 UTC 2024 - 284.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/ir/tfl_ops.td
: OpBuilder::atBlockEnd(body, listener); } }]; // Canonicalizer wasn't defined for this one. In practise, we legalize the // tf.IfOp to scf.If op first and then legalize it to tfl.if to reduce // code redundancy. } def TFL_WhileOp : Op<TFL_Dialect, "while", [ DeclareOpInterfaceMethods<LoopLikeOpInterface, ["isDefinedOutsideOfLoop"]>,
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Jun 06 19:09:08 UTC 2024 - 186K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/transforms/legalize_variables.cc
namespace mlir { namespace TFL { namespace { #define GEN_PASS_DEF_LEGALIZEVARIABLESPASS #include "tensorflow/compiler/mlir/lite/transforms/passes.h.inc" // Attribute name to identify whether variables should be legalized to TFLite or // not. const char kLegalizeTflVariables[] = "tfl._legalize_tfl_variables"; bool HasSupportedElementType(Operation* op) { return utils::IsSupportedVariableType(op); }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 3.3K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/api/v1/compile_mlir_util.h
// . Legalizes control flow operations. // . Decomposes compound resource operations so that the only remaining // operations on resource variables are resource reads/writes.. // . Replaces resource reads/writes with function inputs/outputs and // eliminates the use of resource variables. // . Legalizes the operations to XLA HLO operations.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 21 17:24:39 UTC 2024 - 10.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/g3doc/_includes/tf_passes.md
After serialization, this pass removes the `_entry_function` attribute from `tf.XlaCallModule`, and removes all the serialized stablehlo functions from the top-level module. ### `-tfe-legalize-tfg`
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Aug 02 02:26:39 UTC 2023 - 96.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/transforms/legalization_op_config.h
#include "mlir/Support/TypeID.h" // from @llvm-project namespace mlir { namespace mhlo { // Given the Operation, return True if the op is legalized with MLIR. // Returns false if it is legalized with an old TF2XLA Kernel. bool IsOpLegalizedWithMlir(Operation& op); // Given the type ID, check if it's legalized with MLIR. bool IsTypeLegalizedWithMlir(const TypeID& type_id); // Returns true if the op is considered a dynamic padder op.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Feb 26 20:53:25 UTC 2024 - 2K bytes - Viewed (0) -
tensorflow/compiler/mlir/lite/flatbuffer_export.cc
Value value, const std::string& name, unsigned buffer_idx, const std::optional<BufferOffset<tflite::QuantizationParameters>>& quant_parameters); // TODO(b/137395003): Legalize tf.IfOp to TFLite dialect, and change the // following method to handle TFL::IfOp. BufferOffset<tflite::Operator> BuildIfOperator( mlir::TF::IfOp op, const std::vector<int32_t>& operands,
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Jun 12 21:41:49 UTC 2024 - 164.5K bytes - Viewed (2) -
tensorflow/compiler/mlir/lite/stablehlo/transforms/transforms.h
bool smuggle_disallowed_ops); // This function is a common entry point for all graph optimizations that are // not specific to any hardware. It legalizes SHLO->MHLO, does MHLO->MHLO // optimizations by calling `AddMhloOptimizationPasses` internally, and // legalizes MHLO->SHLO void AddStablehloOptimizationPasses(OpPassManager& pm); // Adds all the backend-agonstic stableHLO optimization passes
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Feb 21 01:08:27 UTC 2024 - 1.9K bytes - Viewed (0)