- Sort Score
- Result 10 results
- Languages All
Results 11 - 20 of 326 for computation (0.17 sec)
-
tensorflow/compiler/mlir/tensorflow/utils/tf_xla_mlir_translate.cc
xla::XlaComputation computation, return_value.valid() ? builder.Build(return_value) : builder.Build()); auto hlo_module = computation.proto(); xla::HloProto hlo_proto; hlo_proto.mutable_hlo_module()->Swap(&hlo_module); compilation_result->computation = std::make_shared<xla::XlaComputation>(); xla::XlaComputation* xla_computation = compilation_result->computation.get();
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 18.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/api/v1/compile_tf_graph.h
namespace v1 { // Compiles the given Tensorflow graph into xla::HLO. The result is in // compilation_result. If the input computation is in MLIR, it will be // converted to a Tensorflow graph. Otherwise, the graph compiler will be run. absl::Status CompileTensorflowGraphToHlo( const std::variant<tpu::MlirToHloArgs, tpu::FunctionToHloArgs>& computation, const tpu::TPUCompileMetadataProto& metadata, bool use_tuple_args,
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Sat Apr 13 08:08:57 UTC 2024 - 2.1K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/transforms/tf2xla_rewriter.cc
XlaComputation& computation) { xla::DebugOptions debug_options; TF_ASSIGN_OR_RETURN(auto hlo_module_config, xla::HloModule::CreateModuleConfigFromProto( computation.proto(), debug_options)); TF_ASSIGN_OR_RETURN( std::unique_ptr<xla::HloModule> hlo_module, xla::HloModule::CreateFromProto(computation.proto(), hlo_module_config));
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu May 02 09:16:07 UTC 2024 - 18.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/internal/passes/clustering_passes.h
CreateTPUClusterFormationPass(bool strict_clusters = false); // Creates a pass that extracts outside compilation (Host ops inside device // cluster) at head/tail of Device cluster to run before/after XLA computation. std::unique_ptr<mlir::OperationPass<mlir::ModuleOp>> CreateExtractHeadTailOutsideCompilationPass(); // Creates a pass that extract outside compilation (Host ops inside cevice // cluster) ops to a separate parallel_execute region to run on CPU.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Apr 30 02:01:13 UTC 2024 - 3.5K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/internal/passes/extract_head_tail_outside_compilation.cc
} // Returns a set of ops that are outside compiled and can be extracted to before // the TPU computation. These ops are either connected to the inputs of the TPU // computation or other ops that can be extracted, and have no operands from // other ops in the TPU computation that cannot be extracted. llvm::SmallVector<Operation*, 4> FindOutsideCompiledOpsAtHead(
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Apr 25 16:01:03 UTC 2024 - 20.6K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/api/v2/legalize_tf.h
// arg_core_mapping - Which args go on which cores. // per_core_arg_shapes - For each core, the shapes for each argument. // client - The Xla Compilation client. absl::StatusOr<tensorflow::XlaCompilationResult> LegalizeMlirToHlo( const std::variant<tpu::MlirToHloArgs, tpu::FunctionToHloArgs>& computation, const tpu::TPUCompileMetadataProto& metadata, bool use_tuple_args, llvm::StringRef device_type,
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Sat Apr 13 07:32:57 UTC 2024 - 2.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/internal/legalize_tf_mlir.h
// then those passes include all the Legalization to XLA HLO which is returned // in the compilation_result. absl::StatusOr<std::string> CompileFromMlirToXlaHlo( bool lower_to_xla_hlo, const tpu::MlirToHloArgs& computation, const tpu::TPUCompileMetadataProto& metadata, llvm::StringRef device_type, const XlaShapeLayoutHelpers::ShapeDeterminationFns& shape_determination_fns,
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Sun Apr 14 20:29:34 UTC 2024 - 2.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/tests/tpu_rewrite.mlir
%add_result = "tf.Add"(%computation, %computation) : (tensor<i32>, tensor<i32>) -> tensor<i32> // expected-error@+1 {{Input of TPUPartitionedOutputV2 must be in tpu computation.}} %partitioned_output:2 = "tf.TPUPartitionedOutputV2"(%add_result) {N = 2 : i64, partition_dims = []} : (tensor<i32>) -> (tensor<i32>, tensor<i32>)
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu May 02 22:03:30 UTC 2024 - 172.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/api/v1/compile_mlir_util.h
// . Legalizes the operations to XLA HLO operations. // . Canonicalizes the XLA HLO operations. // // device_type: XLA JIT device to use for compilation such as "XLA_CPU_JIT", // "XLA_GPU_JIT" or "XLA_TPU_JIT". // use_tuple_args: when this is true, always create a tuple argument for the // entry computation. // enable_op_fallback: when this is true, prefer tf2xla fallback kernels over // MLIR // native kernels for legalization to HLO.
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/tf2xla/internal/passes/extract_outside_compilation.cc
} } } // Since we have the outputs from host and device computation after moving // outside compiled ops, we can create the actual parallel_execute regions. // Still, one region is for the host computation for outside compilation and // the other one is for the original Device cluster computation. mlir::tf_device::ParallelExecuteOp CreateFinalParallelExecuteOp(
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Apr 30 21:25:12 UTC 2024 - 68.3K bytes - Viewed (0)