0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 3 years have passed since last update.

TF &Torch version for GPU CUDA

0
Posted at
  • 互換性 (22.10)

    • MLやRL 開発の Frameworkとして、Tensorflow 2.X, Torchをよく使用するが、GPU版を実装するときにはversion 互換性を注意すべきである。うまく動作しない場合も存在する
  • PyTorch

version cuda remarks
1.13 11.7, 11.6
1.12 11.6, 11.3, 10.2
1.11 11.3, 10.2
version numpy remarks
2.8 ~ >= 1.20
2.7 >= 1.14.5 keras 2.6 needs 1.19.2, h5py needs 1.14.5 for py37
2.4,2.5,2.6 ~=1.19.2
2.2, 2.3 >= 1.16.0, < 1.19.0

最新版TF2.11: tf2.10, cuda 11.8 cudnn 8.6(22.10) numpy >= 1.20
TensorFlow 2.5 の GPU 版での,cuDNN のバージョンは 8.2
TensorFlow 2.4 の GPU 版での,cuDNN のバージョンは 8.0.5
TensorFlow 2.3, 2.2, 2.1 の GPU 版での,cuDNN のバージョンは 7.6
そして,TensorFlow 1.13 以上 TensorFlow 2.0 までの GPU 版での,cuDNN のバージョンは7.4 が良い

※about detail: http://tecsingularity.com/cuda/version-2/

0
0
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?