Sign inSign up

gperdrizet/deeplearning-gpu

By gperdrizet

•Updated 4 months ago
Archived

Containerized deeplearning GPU development environment

Image
Machine learning & AI
Developer tools
Data science
1

2.6K

gperdrizet/deeplearning-gpu repository overview

⁠Deep learning GPU development environment

A ready-to-use deep learning environment with NVIDIA GPU support for VS Code. Includes both PyTorch and TensorFlow frameworks. Designed for cross-platform support and wide GPU compatibility.

Base image: NVIDIA TensorFlow 25.02⁠

Note: This container is designed to be used as part of a VS Code Dev Container environment, not run directly. See the full environment setup on GitHub⁠ for devcontainer configuration and usage instructions.

CUDA version note: This image uses CUDA 12.8, which supports GPU architectures from Pascal (GTX 10xx) through Blackwell (RTX 50xx/B100/B200). CUDA 12.x is the last series to support Pascal; CUDA 13.0 removed Pascal, Maxwell, and Volta support entirely⁠. CUDA 12.8 requires host driver ≥ 570.x on Linux. Note: 25.02 is the final NVIDIA TensorFlow NGC container release.

⁠Features

  • Dual Framework Support – Both TensorFlow 2.17 and PyTorch 2.11 pre-installed and GPU-ready
  • NVIDIA GPU Acceleration – Built on NVIDIA's official TensorFlow container with CUDA 12.8 and Blackwell support
  • VS Code Dev Container Ready – Pre-configured vscode user for seamless devcontainer integration

⁠Included software

ComponentVersion
Base Imagenvcr.io/nvidia/tensorflow:25.02-tf2-py3
TensorFlow2.17
Keras3.x
PyTorch2.11 (custom build, CUDA 12.8)
Python3.12
CUDA12.8
GPU SupportPascal through Blackwell (sm_60 - sm_100)
⁠Additional Python packages
  • matplotlib - Data visualization
  • seaborn - Statistical data visualization
  • keras_tuner - Hyperparameter tuning for Keras
  • optuna - General purpose hyperparameter optimization framework
  • ipywidgets - Interactive widgets for Jupyter

⁠Usage with VS Code dev containers

This image is optimized for use with VS Code Dev Containers. The vscode user (UID 1000) is pre-configured with sudo access for a seamless development experience.

⁠Running directly with Docker

You can also run the container directly with Docker and launch a Jupyter server:

docker run --gpus all -p 8888:8888 -v $(pwd):/workspace gperdrizet/deeplearning-gpu:latest \
    jupyter notebook --ip=0.0.0.0 --port=8888 --no-browser --allow-root --notebook-dir=/workspace \
    --NotebookApp.token=''

This will:

  • Enable GPU access with --gpus all
  • Map port 8888 for Jupyter access
  • Mount your current directory to /workspace in the container
  • Start a Jupyter notebook server accessible at http://localhost:8888
  • Disable token authentication (suitable for LAN-only environments)

Security Note: Only disable authentication on trusted networks. For public or untrusted networks, remove --NotebookApp.token='' to require token authentication.

⁠Requirements

  • Docker with NVIDIA GPU support
  • NVIDIA GPU with compatible drivers

⁠License

See the GitHub repository⁠ for license information.

Tag summary

Content type

Image

Digest

sha256:a5d3bd16b…

Size

9.9 GB

Last updated

4 months ago

docker pull gperdrizet/deeplearning-gpu