- Sort Score
- Result 10 results
- Languages All
Results 1 - 6 of 6 for computations (0.15 sec)
-
tensorflow/compiler/mlir/tensorflow/transforms/passes.h
// Creates a pass that lifts operations on external resource variables from // device computation nested in `tf_device::LaunchOp` out so that resource // variable load operations are all before device computation while resource // variable store operations are all after device computation. After this pass, // device computation no longer interacts with external resource variables.
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/tf2xla/api/v1/compile_tf_graph.cc
} if (output_to_input_alias.empty()) return absl::OkStatus(); xla::HloModuleProto* module_proto = compilation_result->computation->mutable_proto(); absl::StatusOr<xla::ProgramShape> program_shape_or_status = compilation_result->computation->GetProgramShape(); TF_RET_CHECK(program_shape_or_status.ok()); xla::ProgramShape& program_shape = program_shape_or_status.value();
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Jun 12 22:19:26 UTC 2024 - 14K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/utils/tpu_rewrite_device_util.cc
} return DeviceNameUtils::ParsedNameToString(tpu_device); } // Determine execution devices when topology and device assignment are not // defined. This is a special case where a single core computation is replicated // to every core in the mesh. TPU devices are simply added to // `execution_devices` of one replica. `num_replicas` must be 1 or the total // number of TPU devices available, and `num_cores_per_replica` must be 1.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Jun 10 20:10:40 UTC 2024 - 32.8K bytes - Viewed (0) -
src/cmd/compile/internal/liveness/plive.go
// // OpVarDef x // base, len, cap = pieces of x[1:] // x = {base, len, cap} // // then the liveness analysis would decide the previous value of x was // unnecessary even though it is about to be used by the x[1:] computation. // Similarly, if the generated code looked like: // // base, len, cap = pieces of x[1:] // x = {base, len, cap} // OpVarDef x // // then the liveness analysis will not preserve the new value of x, because
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Fri Jun 07 15:22:22 UTC 2024 - 45.2K bytes - Viewed (0) -
android/guava-tests/test/com/google/common/util/concurrent/JSR166TestCase.java
* SMALL_DELAY_MS}, {@code MEDIUM_DELAY_MS}, {@code LONG_DELAY_MS}. The idea here is that a * SHORT is always discriminable from zero time, and always allows enough time for the small * amounts of computation (creating a thread, calling a few methods, etc) needed to reach a * timeout point. Similarly, a SMALL is always discriminable as larger than SHORT and smaller
Registered: Wed Jun 12 16:38:11 UTC 2024 - Last Modified: Mon Jun 10 19:21:11 UTC 2024 - 37.7K bytes - Viewed (0) -
src/math/big/float.go
rbit := z.mant.bit(r) & 1 // rounding bit; be safe and ensure it's a single bit // The sticky bit is only needed for rounding ToNearestEven // or when the rounding bit is zero. Avoid computation otherwise. if sbit == 0 && (rbit == 0 || z.mode == ToNearestEven) { sbit = z.mant.sticky(r) } sbit &= 1 // be safe and ensure it's a single bit // cut off extra words
Registered: Wed Jun 12 16:32:35 UTC 2024 - Last Modified: Thu Jun 06 15:46:54 UTC 2024 - 44.5K bytes - Viewed (0)