It’s a great time to create a project with DataRobot! Everything you do in DataRobot starts with creating a project; and then, our workers step in to help out. I know what you’re thinking: workers are everywhere, they’re you and me! Well, in DataRobot “workers” are a little different—and also, they’re not human. DataRobot leverages “little data scientists” under the hood when training models or creating insights. These workers manage the queue of modeling requests (across all your projects) and streamline processes. (For all the detailed information about workers, see the documentation.) In this tip, I want to make sure you understand what you can expect when using the AI Cloud Platform Trial.
In general, you can increase and decrease the number of workers allocated for a project, right here:
(By the way, for the project in this image you can only decrease workers as indicated by the orange down arrow. Also, this project is only using 3 of 8 workers because I was building models for multiple projects at the same time—which is so cool.)
Each project created with AI Cloud Platform Trial is allocated with the maximum number of workers, right out of the box. So you can sit back and relax knowing that DataRobot and its little data scientists are chugging away as fast and efficiently as possible to build the best models!
This prompts the question: Does increasing or decreasing the number of workers affect credits used? No, it does not! The overall amount of work remains the same. So, there’s no need to consider decreasing workers to save credits.
It’s an amazing thing to be able to allocate virtual data scientists to look into your data and produce better and better models over time. If you’re using the AI Cloud Platform Trial and want to learn a bit more about workers, you can check out the Build AI Models Tour which has a section illustrating workers and where they’re used.
Need help finding tours? Check out this other fast Tip of the Day!
Looking forward to more of your ideas! Remember to comment below or DM me, @Cyant_DR.
Hi Cyant-DR - this is great -thank you for explaining. It makes me want to ask how many projects can I build at the same time? What's the limit? no rush on your answer
@mikio note there's no limit to concurrent projects - but you'll only have so many available workers to process the tasks within them, as they are shared across your workload. To maximize throughput you'll want to keep your workers as busy as possible - but to reduce time to project completion for any particular project, you'll either want to focus on one or two projects at a time, or look into purchasing additional workers.