Setting up Dotscience Runners
This section describes how to attach your machine (runner) to a Dotscience install (which itself can either hosted by us or your own organization). In our system the user supplies the compute power (own machine, cloud, VM, GPU, etc.), which carries great flexibility but means information about where the compute is located is needed (unless you are using the Dotscience-provided runner).
Installation and set up
1. Create a user account for the cloud platform
Collaboration features, performance visualisations, version histories, and other information about your tracked models can be accessed via the Dotscience cloud platform in your web browser. To start, go to cloud.dotscience.com and sign up.
Or, if your organization is using a self-hosted Dotscience setup, log into the Dotscience setup provided by your organization.
2. Set up your runner
When you use Dotscience, your models run on a designated machine called a runner. This means that the underlying compute will be handled by whichever machine you choose as the best fit for your needs, including powerful cloud instances or GPUs. You can use (just about) any computer as the runner: your laptop or local server, or a VM on the cloud. You need to tell Dotscience about your runner so that it can track and version your models.
Every new account in Dotscience Cloud will auto-provision a new runner, which will be shut down after 2 hours of inactivity. When you create a new task (i.e start Jupyter, run a python script or build a new model), runners will auto provision again, ready to be used.
If you want custom runners, pick a runner setup guide from this section corresponding to your runner location. If you want to use a cloud VM as your runner, the guides in this section will point you in the right direction to set one up.