Skip to content

deeplearning images

Full-featured deep learning environment with TensorFlow and PyTorch. Three variants: NVIDIA GPU, CPU-only, and Apple Silicon.


deeplearning-nvidia

Based on NVIDIA's NGC TensorFlow container, which provides a validated CUDA + cuDNN + TensorFlow stack with Pascal support out of the box. PyTorch and CuPy are added via custom-built wheels (see Building Custom Wheels).

Component Version
Base image nvcr.io/nvidia/tensorflow:25.02-tf2-py3
Python 3.12
CUDA 12.8
TensorFlow 2.17
PyTorch 2.11.0 (custom wheel)
CuPy 13.6.0 (custom wheel)
Keras 3.x
GPU support Pascal-Blackwell (sm_60-sm_100)

Other packages: numpy, pandas, scikit-learn, scipy, matplotlib, seaborn, plotly, jupyterlab, keras_tuner, optuna, tensorboard, python-dotenv

DockerHub: gperdrizet/deeplearning-nvidia


deeplearning-cpu

Component Version
Base image python:3.12-slim
Python 3.12
TensorFlow 2.x (latest via pip)
PyTorch Latest CPU (via pip)

Other packages: numpy, pandas, scikit-learn, scipy, matplotlib, seaborn, plotly, jupyterlab, keras_tuner, optuna, tensorboard, python-dotenv

DockerHub: gperdrizet/deeplearning-cpu


deeplearning-mac

Native linux/arm64 image for Apple Silicon (M1/M2/M3). No Rosetta emulation. GPU acceleration is not available inside Docker on macOS.

Component Version
Base image python:3.12-slim
Platform linux/arm64
Python 3.12
TensorFlow 2.17
Keras 3.x
PyTorch Latest (CPU, ARM64)

Other packages: numpy, pandas, scikit-learn, scipy, matplotlib, seaborn, plotly, jupyterlab, keras_tuner, optuna, tensorboard, python-dotenv

DockerHub: gperdrizet/deeplearning-mac