Getting Started¶
Installation¶
Use pip to install or upgrade Spotty:
pip install -U spotty
Python >=3.6 is required.
Also, depending on the use case, some additional software is needed:
Docker if you want to run containers locally: Get Docker
AWS CLI if you're going to use AWS: Installing the AWS Command Line Interface
Google Cloud SDK if you're going to use GCP: Installing Google Cloud SDK
Prepare a configuration file¶
Prepare a spotty.yaml
file and put it to the root directory of your project:
See the file specification and an example here: Spotty Configuration File.
Read this article for a real-world example.
Start an instance¶
Use the following command to launch an instance with the Docker container:
spotty start
If you're using AWS, it will create EBS volumes if needed, start an instance, upload project files and start the Docker container with the environment for your project.
Train your models or run notebooks¶
To connect to the running container via SSH, use the following command:
spotty sh
It runs a tmux session, so you can always detach this session using
Ctrl + b
, then d
combination of keys. To be attached to that session later, just use the
spotty sh
command again.
Also, you can run custom scripts inside the Docker container using the spotty run <SCRIPT_NAME>
command. Read more
about custom scripts in the documentation: Configuration File: "scripts" section.