behzadhakki/magenta_paperspace repository overview

⁠Installation

⁠1. Go to the following DockerHub Repo

https://hub.docker.com/repository/docker/behzadhakki/magenta_paperspace

⁠2. Pull the latest version or an earlier version

docker pull behzadhakki/magenta_paperspace:0.2

⁠3A. Run on CPU (command line mode)

⁠Without access to local computer

docker run -it --rm behzadhakki/magenta_paperspace:0.2 bash

⁠With access to local computer (i.e. Run the Docker image with access to local computer, local computer mounted in the “mycomputer” folder)

docker run -v ~:/tf/mycomputer -it --rm behzadhakki/magenta_paperspace:0.2 bash

⁠or 3B. Run on CPU (Jupyter notebook mode)

⁠Without access to local computer

docker run -it -p 8888:8888 behzadhakki/magenta_paperspace:0.2 

⁠With access to local computer (i.e. Run the Docker image with access to local computer, local computer mounted in the “mycomputer” folder)

docker run -v ~/:/tf/mycomputer -it -p 8888:8888 behzadhakki/magenta_paperspace:0.2 

⁠Note: to run on GPU, add the following argument to the docker command

    --runtime=nvidia

⁠4. Activate magenta miniconda environment if it is not active

. /usr/local/etc/profile.d/conda.sh
conda activate magenta

⁠NOTE: the source code can be found here: /usr/local/envs/magenta/lib/python3.6/site-packages/magenta/

⁠Versions

⁠magenta0

Magenta 1.1.7, Tensorflow1.15

⁠0.1

Visual_midi added to the image

⁠0.2

image added to Git