SuperexponentialAI
/

relu

relu / torch-ext /torch_binding.h

Optimized relu: cpu/cuda/xpu, 1.2-1.85x faster on RTX 4090, benchmarked vs upstream and torch.relu

e873e70 verified about 1 month ago

98 Bytes

	#pragma once

	#include <torch/torch.h>

	void relu(torch::Tensor &out, torch::Tensor const &input);