Deep Learning Containers overview

Deep Learning Containers are a set of Docker containers with key data science frameworks, libraries, and tools pre-installed. These containers provide you with performance-optimized, consistent environments that can help you prototype and implement workflows quickly.

To learn more about containers, see Containers at Google.

Pre-installed software

Deep Learning Containers images can be configured to include the following:

  • Frameworks:

    • TensorFlow
    • PyTorch
    • R
    • scikit-learn
    • XGBoost
  • Python, including these packages:

    • numpy
    • sklearn
    • scipy
    • pandas
    • nltk
    • pillow
    • fairness-indicators for TensorFlow 2.3 and 2.4 Deep Learning Containers instances
    • many others
  • Nvidia packages with the latest Nvidia driver for GPU-enabled instances:

    • CUDA 10.* and 11.* (the version depends on the framework)
    • CuDNN 7.* and NCCL 2.* (the version depends on the CUDA version)
  • JupyterLab

What's next

You can get started with Deep Learning Containers by walking through the How-to guides, which provide instructions on create and work with deep learning containers.