- Sort Score
- Result 10 results
- Languages All
Results 1 - 10 of 10 for TPUPartitionedInput (0.33 sec)
-
tensorflow/compiler/mlir/tensorflow/tests/tpu_partitioned_op_conversion.mlir
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri Jan 20 17:43:51 UTC 2023 - 8.3K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/tests/executor_tpuv1_island_coarsening/executor_tpuv1_island_coarsening.mlir
// Check that we bring in TPUPartitionedInput operand producers. // CHECK-LABEL: func @fuse_in_partitioned_input_op func.func @fuse_in_partitioned_input_op(%arg0: tensor<2x4xf32>, %arg1: tensor<2x4xf32>) { tf_executor.graph { // CHECK: tf_executor.island // CHECK-NEXT: = "tf.TPUPartitionedInput" // CHECK-NEXT: = "tf.Const" // CHECK-NEXT: = "tf.AddV2"
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Aug 02 03:15:59 UTC 2022 - 36.2K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/internal/passes/tpu_sharding_identification_pass.cc
mlir::Operation* NullUnlessSharded(PartitionedOp op) { return op.get_XlaSharding() ? op : nullptr; } // Returns a TPUPartitionedInput op connected to a `tf_device.cluster_func` // operand value if it has an XLA sharding. If value is a resource type then // TPUPartitionedInput op will be connected to a ReadVariable op that feeds into // a `tf_device.cluster_func`. mlir::Operation* GetXlaShardingFromOperand(Value value) {
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Apr 30 02:01:13 UTC 2024 - 28.9K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/tests/tpu_validate_inputs.mlir
// expected-error @+1 {{'tf.TPUPartitionedInput' op TF2XLA TPU bridge input check: number of inputs inconsistent. num_cores_per_replica=2 no. of inputs=3}} %pi, %c0 = tf_executor.island wraps "tf.TPUPartitionedInput"(%arg0, %arg1, %arg1) {index = 1 : i64} : (tensor<i32>, tensor<i32>, tensor<i32>) -> tensor<i32>
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue May 07 06:51:01 UTC 2024 - 15.7K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/transforms/executor_tpuv1_island_coarsening.cc
std::vector<IslandOp> child_ops; if (incoming) { // Looks at captured operands of `candidate_wrapped_op` to bring special // TPU ops such as tf.TPUReplicatedInput and tf.TPUPartitionedInput into // the island as well. These ops are brought in only if they do not // already have a cluster assigned to them (via `_replication_info` // attribute value).
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri May 17 17:58:54 UTC 2024 - 27.6K bytes - Viewed (0) -
tensorflow/compiler/mlir/g3doc/_includes/tf_passes.md
inputs. It reorders `tf.TPUPartitionedInput` (model parallelism) and `tf.TPUReplicatedInput` (data parallelism) ops. It transforms a DAG where multiple `tf.TPUPartitionedInput` ops are feeding into a single `tf.TPUReplicatedInput` into a DAG where multiple `tf.TPUReplicatedInput` ops are feeding into a single `tf.TPUPartitionedInput`. Transforming the IR in such
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/tensorflow/transforms/tf_passes.td
inputs. It reorders `tf.TPUPartitionedInput` (model parallelism) and `tf.TPUReplicatedInput` (data parallelism) ops. It transforms a DAG where multiple `tf.TPUPartitionedInput` ops are feeding into a single `tf.TPUReplicatedInput` into a DAG where multiple `tf.TPUReplicatedInput` ops are feeding into a single `tf.TPUPartitionedInput`. Transforming the IR in such
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Jun 12 21:18:05 UTC 2024 - 99.6K bytes - Viewed (0) -
tensorflow/compiler/mlir/tf2xla/internal/passes/clustering_passes.td
let description = [{ Bubbles up sharding configuration from `cluster_func` regions into the attributes of `cluster_func`. This is done by parsing the `XlaSharding` / `TPUPartitionedOutput` / `TPUPartitionedInput` ops inside `cluster_func`. For example, given the following `cluster_func` wrapping `func`: ```mlir func @test(%arg0: tensor<*xi32>) { "tf_device.cluster_func"(%arg0) {
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Apr 30 02:01:13 UTC 2024 - 19.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/transforms/passes.h
// attributes back to legacy attributes. std::unique_ptr<OperationPass<func::FuncOp>> CreateConvertToLegacyCompileAndReplicateAttributesPass(); // Creates a pass that converts all TPUPartitionedInput to TPUPartitionedInputV2 std::unique_ptr<OperationPass<func::FuncOp>> CreateTPUPartitionedOpConversionPass(); std::unique_ptr<OperationPass<ModuleOp>> CreateTPUValidateInputsPass();
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Jun 12 21:18:05 UTC 2024 - 31.8K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/ir/tf_ops.td
TF_Int64Tensor:$device_ordinal ); } // We must manually define TPUPartitionedInput, TPUPartitionedInputV2, // TPUPartitionedOutput, and TPUPartitionedOutputV2 since they have an // optional attribute, _XlaSharding, unlike the TensorFlow definition. def TF_TPUPartitionedInputOp : TF_Op<"TPUPartitionedInput", [Pure]> { let summary = [{ An op that groups a list of partitioned inputs together. This op
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Apr 24 04:08:35 UTC 2024 - 90.5K bytes - Viewed (0)