Docker image for tensorflow
You can get the image from the Docker Hub or you can build it with the Dockerfile
You can pull the image from the Docker Hub
$ docker pull orianna/tensorflow
- 1.5.0-py3.6 (tensorflow-1.5.0, python 3.6, anaconda3-5.0.1)
- 1.13.1, latest (tensorflow-1.13.1, python 3.6, anaconda3-5.2.0, add keras)
You can build it from the Dockerfile. Base image: orianna/tensorflow
$ docker build -t your-image-name:your-tag .
Working directory is /work
. You can mount working directory to your code directory on your host. You can go to your code directory and use the following command to start the container:
$ docker run -v $(pwd):/work -p 8888:8888 -p 6006:6006 --rm -it orianna/tensorflow
After you start your container, you can use the following command to start the Jupyter Notebook:
$ jupyter notebook --ip=0.0.0.0 --allow-root --NotebookApp.token=''
Now you can open your browser with 0.0.0.0:8888 and enjoy notebook!
You can also use tensorboard with the this command.
$ tensorboard --logdir $LOG_DIR --host 0.0.0.0