tonyhoo / docker-autogluon Goto Github PK

Install Instructions | Documentation (Stable | Latest)

AutoGluon automates machine learning tasks enabling you to easily achieve strong predictive performance in your applications. With just a few lines of code, you can train and deploy high-accuracy machine learning and deep learning models on image, text, time series, and tabular data.

Visit auto.gluon.ai for more information.

• framework - base framework image - contains AutoGluon, runtimes and necessary dependencies
• jupyter - image with a jupyter notebook available

0.8.2-cpu-framework-ubuntu20.04-py3.9
 ^     ^   ^          ^          ^
 |     |   type       |          python version
 |     |              |
 |     |              |
 |     |              |
 |     CPU type       linux version
 |
 AutoGluon version

0.8.2-cuda11.7-jupyter-ubuntu20.04-py3.9
 ^        ^    ^          ^          ^
 |        |    type       |          python version
 |        |               |
 |        cuda version    |
 |                        |
 |                        linux version
 |
 AutoGluon version

Start Container and Notebook Server with GPU support

$ docker pull autogluon/autogluon:0.8.2-cuda11.7-jupyter-ubuntu20.04-py3.9
$ docker run --gpus all --shm-size=1G --rm -it -p 8888:8888 \
         autogluon/autogluon:0.8.2-cuda11.7-jupyter-ubuntu20.04-py3.9

Start Container and Notebook Server with CPU-only support

$ docker pull autogluon/autogluon:0.8.2-cpu-jupyter-ubuntu20.04-py3.9
$ docker run --rm --shm-size=1G -it -p 8888:8888 \
         autogluon/autogluon:0.8.2-cpu-jupyter-ubuntu20.04-py3.9