From 3ebcc647b651a4a3c717eff2a3087127e5707e0c Mon Sep 17 00:00:00 2001
From: AlexeyAB <alexeyab84@gmail.com>
Date: Mon, 14 May 2018 09:20:38 +0000
Subject: [PATCH] Fixed network resizing (random=1) for non-square networks
---
src/convolutional_layer.c | 5 ++++-
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/src/convolutional_layer.c b/src/convolutional_layer.c
index 91c5b3c..b8065fd 100644
--- a/src/convolutional_layer.c
+++ b/src/convolutional_layer.c
@@ -178,6 +178,8 @@
// batch norm
cudnnSetTensor4dDescriptor(l->normTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, 1, l->out_c, 1, 1);
cudnnSetTensor4dDescriptor(l->normDstTensorDesc, CUDNN_TENSOR_NCHW, CUDNN_DATA_FLOAT, l->batch, l->out_c, l->out_h, l->out_w);
+
+ cudnnSetTensor4dDescriptor(l->normDstTensorDescF16, CUDNN_TENSOR_NCHW, data_type, l->batch, l->out_c, l->out_h, l->out_w);
#if(CUDNN_MAJOR >= 6)
cudnnSetConvolution2dDescriptor(l->convDesc, l->pad, l->pad, l->stride, l->stride, 1, 1, CUDNN_CROSS_CORRELATION, CUDNN_DATA_FLOAT); // cudnn >= 6.0
#else
@@ -379,6 +381,7 @@
}
#ifdef CUDNN
cudnnCreateTensorDescriptor(&l.normDstTensorDesc);
+ cudnnCreateTensorDescriptor(&l.normDstTensorDescF16);
cudnnCreateTensorDescriptor(&l.normTensorDesc);
cudnnCreateTensorDescriptor(&l.srcTensorDesc);
cudnnCreateTensorDescriptor(&l.dstTensorDesc);
@@ -488,7 +491,7 @@
size_t total_byte;
check_error(cudaMemGetInfo(&free_byte, &total_byte));
if (l->workspace_size > free_byte || l->workspace_size >= total_byte / 2) {
- printf(" used slow CUDNN algo without Workspace! \n");
+ printf(" used slow CUDNN algo without Workspace! Need memory: %d, available: %d\n", l->workspace_size, (free_byte < total_byte/2) ? free_byte : total_byte/2);
cudnn_convolutional_setup(l, cudnn_smallest);
l->workspace_size = get_workspace_size(*l);
}
--
Gitblit v1.10.0