Spotty drastically simplifies training of deep learning models on AWS:
- it makes training on AWS GPU instances as simple as training on your local computer
- it automatically manages all necessary AWS resources including AMIs, volumes, snapshots and SSH keys
- it makes your model trainable on AWS by everyone with a couple of commands
- it uses tmux to easily detach remote processes from their terminals
- it saves you up to 70% of the costs by using Spot Instances
- See the documentation page.
- Read this article on Medium for a real-world example.
Requirements:
- Python >=3.5
- AWS CLI (see Installing the AWS Command Line Interface)
Use pip to install or upgrade Spotty:
$ pip install -U spotty
-
Prepare a
spotty.yaml
file and put it to the root directory of your project: -
Start an instance:
$ spotty start
It will run a Spot Instance, restore snapshots if any, synchronize the project with the running instance and start the Docker container with the environment.
-
Train a model or run notebooks.
To connect to the running container via SSH, use the following command:
$ spotty ssh
It runs a tmux session, so you can always detach this session using
Ctrl + b
, thend
combination of keys. To be attached to that session later, just use thespotty ssh
command again.Also, you can run your custom scripts inside the Docker container using the
spotty run <SCRIPT_NAME>
command. Read more about custom scripts in the documentation: Configuration: "scripts" section.
Any feedback or contributions are welcome! Please check out the guidelines.