Skip to content

Commit be42a12

Browse files
committed
Update INSTALL.md
1 parent a6bc063 commit be42a12

File tree

1 file changed

+5
-5
lines changed

1 file changed

+5
-5
lines changed

INSTALL.md

+5-5
Original file line numberDiff line numberDiff line change
@@ -243,14 +243,14 @@ cuDNN is a library from NVIDIA that efficiently implements many of the operation
243243
that are commonly used in deep learning.
244244

245245
After registering as a developer with NVIDIA, you can [download cuDNN here](https://developer.nvidia.com/cudnn).
246+
Make sure to download Version 4.
246247

247248
After dowloading, you can unpack and install cuDNN like this:
248249

249-
```
250-
tar -xzvf cudnn-6.5-linux-x64-v2.tgz
251-
cd cudnn-6.5-linux-x64-v2/
252-
sudo cp libcudnn* /usr/local/cuda-7.0/lib64
253-
sudo cp cudnn.h /usr/local/cuda-7.0/include
250+
```bash
251+
tar -xzvf cudnn-7.0-linux-x64-v4.0-prod.tgz
252+
sudo cp cuda/lib64/libcudnn* /usr/local/cuda-7.0/lib64/
253+
sudo cp cuda/include/cudnn.h /usr/local/cuda-7.0/include/
254254
```
255255

256256
Next we need to install the torch bindings for cuDNN:

0 commit comments

Comments
 (0)