From 1b2c70f82a9ec88c60bb277f9f5f0073c96691e7 Mon Sep 17 00:00:00 2001
From: AlexeyAB <alexeyab84@gmail.com>
Date: Tue, 03 Apr 2018 13:42:00 +0000
Subject: [PATCH] Detection is accelerated by 7 percent (fused conv and batch_norm layers)
---
src/maxpool_layer_kernels.cu | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/maxpool_layer_kernels.cu b/src/maxpool_layer_kernels.cu
index 6381cc1..d40d3c0 100644
--- a/src/maxpool_layer_kernels.cu
+++ b/src/maxpool_layer_kernels.cu
@@ -92,7 +92,7 @@
size_t n = h*w*c*layer.batch;
- forward_maxpool_layer_kernel<<<cuda_gridsize(n), BLOCK>>>(n, layer.h, layer.w, layer.c, layer.stride, layer.size, layer.pad, state.input, layer.output_gpu, layer.indexes_gpu);
+ forward_maxpool_layer_kernel<<<cuda_gridsize(n), BLOCK, 0, get_cuda_stream()>>>(n, layer.h, layer.w, layer.c, layer.stride, layer.size, layer.pad, state.input, layer.output_gpu, layer.indexes_gpu);
check_error(cudaPeekAtLastError());
}
--
Gitblit v1.10.0