#123123rd of 126 in ML frameworks and libraries
deepo
Dockerfile generator and prebuilt images for deep learning environments
- Stars
- 6.3k
- License
- MIT
- Last commit
- Mar 2026
- Last release
- Nov 2017
- Language
- Python
6,260 on 4 Apr 20266,281 on 31 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Deepo generates Dockerfiles for deep learning environments from one command, resolving compatible CUDA, cuDNN, Python and framework versions and ordering installs by topological sort. It also publishes prebuilt Docker Hub images: an all-in-one image with TensorFlow, PyTorch, Keras, MXNet, Chainer, Darknet and PaddlePaddle, plus per-framework tags for CUDA 11.3 and CPU-only.
Who it is for: ML researchers who want reproducible Docker-based framework environments
Strengths
- Generator resolves CUDA, cuDNN, Python and framework compatibility and install order automatically
- Per-framework tags let you pull one framework instead of the all-in-one image
- CPU-only images run on Linux, Windows and macOS; GPU images run on Linux
- Jupyter Lab can be launched from the same image on port 8888
Weaknesses
- Current tags target only CUDA 11.3 and Python 3.8, which are old
- Last release is 2017-11-27; Theano, Caffe, CNTK and Torch tags are deprecated
- Custom builds compile some libraries from source and can take several minutes
- GPU use requires the NVIDIA Container Toolkit on the host
What it needs
- GPU optional
- Docker
- Needs Docker, NVIDIA Container Toolkit (GPU only)
- Models: TensorFlow, PyTorch, Keras, MXNet, Chainer
- port 8888
Quick answers
Is deepo open source?
Yes. Its code is on GitHub under the MIT license.
Can you run deepo with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does deepo need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in ML frameworks and libraries
See all 126| Rank | Project | Score |
|---|---|---|
| 1 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 95 out of 100 |
| 2 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |