Deep Learning AMI
Developer Guide

CUDA Installations and Framework Bindings

Deep learning is all pretty cutting edge, however, each framework offers "stable" versions. These stable versions may not work with the latest CUDA or cuDNN implementation and features. How do you decide? This ultimately points to your use case and the features you require. If you are not sure, then go with the latest Deep Learning AMI with Conda. It has official pip binaries of all frameworks with CUDA 8, CUDA 9, and CUDA 10, using whichever most recent version is supported by each framework. If you want the latest versions, and to customize your deep learning environment, go with the Deep Learning Base AMI.

Look at our guide on Stable versus Bleeding Edge for further guidance.

Choosing a DLAMI with CUDA

The Deep Learning Base AMI has CUDA 8, 9 and 10.

The Deep Learning AMI with Conda has CUDA 8 and 9, and 10.

  • CUDA 10 with cuDNN 7: PyTorch

  • CUDA 9 with cuDNN 7: Apache MXNet, Caffe2, Chainer, CNTK, Keras, TensorFlow, Theano

  • CUDA 8 with cuDNN 6: Caffe

For installation options for DLAMI types and operating systems, refer to each of the CUDA version and options pages:

Specific framework version numbers can be found in the DLAMI: Release Notes

Choose this DLAMI type or learn more about the different DLAMIs with the Next Up option.

Choose one of the CUDA versions and review the full list of DLAMIs that have that version in the Appendix, or learn more about the different DLAMIs with the Next Up option.

Next Up

DLAMI Operating System Options

Related Topics