summaryrefslogtreecommitdiffstats
path: root/development/cudnn
diff options
context:
space:
mode:
Diffstat (limited to 'development/cudnn')
-rw-r--r--development/cudnn/README17
1 files changed, 9 insertions, 8 deletions
diff --git a/development/cudnn/README b/development/cudnn/README
index 097b80517b..d49d4ed0b1 100644
--- a/development/cudnn/README
+++ b/development/cudnn/README
@@ -1,12 +1,13 @@
-The NVIDIA CUDA Deep Neural Network library (cuDNN) is a GPU-accelerated
-library of primitives for deep neural networks. cuDNN provides highly tuned
-implementations for standard routines such as forward and backward convolution,
-pooling, normalization, and activation layers. cuDNN is part of the NVIDIA
-Deep Learning SDK.
+The NVIDIA CUDA Deep Neural Network library (cuDNN) is a
+GPU-accelerated library of primitives for deep neural networks. cuDNN
+provides highly tuned implementations for standard routines such
+as forward and backward convolution, pooling, normalization, and
+activation layers. cuDNN is part of the NVIDIA Deep Learning SDK.
-You will need to register for NVIDIA developer account to download the source.
+You will need to register for NVIDIA developer account to download the
+source.
-Make sure you create the CUDA_HOME environment variable and add it to your
-~/.bashrc. The default should be
+Make sure you create the CUDA_HOME environment variable and add it to
+your ~/.bashrc. The default should be
CUDA_HOME=/usr/share/cuda