- Sort Score
- Result 10 results
- Languages All
Results 1 - 1 of 1 for GpuConcatOp (1.68 sec)
-
tensorflow/compiler/mlir/lite/experimental/tac/hardwares/gpu_hardware.cc
} }; std::unique_ptr<TargetHardwareOperation> CreateArithmeticOp() { return std::make_unique<GpuArithmeticOp>(); } // Currently used for these ops: // tfl.concatenation / tfl.reshape class GpuConcatOp : public TargetHardwareOperation { double GetOpCost(mlir::Operation* op) const override { int64_t count; if (ArithmeticCountUtilHelper::GetInputTensorTotalSize(op, &count)) return kGPUCopyUnitCost * count;
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Jun 06 03:08:33 UTC 2023 - 7.8K bytes - Viewed (0)