Test the CUDA Runtime

You can use the following simple examples to test whether the new CUDA ML Runtime is able to leverage GPUs as expected.

  1. Go to a project that is using the CUDA ML Runtime and click New Session.
  2. Launch a new session with GPUs.
  3. Run the following command in the workbench command prompt to verify that the driver was installed correctly:
    ! /usr/bin/nvidia-smi
  4. Use any of the following code samples to confirm that the new engine works with common deep learning libraries.
    PyTorch
    !pip3 install torch
    from torch import cuda
    assert cuda.is_available()
    assert cuda.device_count() > 0
    print(cuda.get_device_name(cuda.current_device()))
    Tensorflow
    !pip3 install tensorflow-gpu==2.1.0
    from tensorflow.python.client import device_lib
    assert 'GPU' in str(device_lib.list_local_devices())
    device_lib.list_local_devices()
    Keras
    !pip3 install keras
    from keras import backend
    assert len(backend.tensorflow_backend._get_available_gpus()) > 0
    print(backend.tensorflow_backend._get_available_gpus())