Skip to content

Commit eb9f858

Browse files
committed
added note on using correct versions of cuda and cudnn given tensorflow version
1 parent eb3e1f7 commit eb9f858

File tree

1 file changed

+2
-0
lines changed

1 file changed

+2
-0
lines changed

README.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -15,6 +15,8 @@ This repository assumes you have an Ubuntu 16.04 x86_64 running Python 3.(5.X).
1515

1616
# GPU Acceleration
1717

18+
> Note: As of current TensorFlow version 1.4.0: `All our prebuilt binaries have been built with CUDA 8 and cuDNN 6. We anticipate releasing TensorFlow 1.5 with CUDA 9 and cuDNN 7.` However, building from source with 1.4.0 using CUDA 9 and cuDNN 7 should be fine and working.
19+
1820
For GPU-accelerated tensorflow, ensure you do or have done the following:
1921

2022
1. You have a GPU with CUDA Compute Capability 3.0 or higher.

0 commit comments

Comments
 (0)