Chris from Optumi here! We are excited to introduce our JupyterLab extension that helps data scientists focus on machine learning experiments without getting bogged down by manual infrastructure work.
Here are just a few reasons people use the tool:
- They want to speed up model training
- They want to try more hyperparameters
- They want to try larger datasets
Here are a few features of the tool:
- Easy access to powerful cloud machines (up to 4 GPU cards, 192 CPU cores, 2 TB of RAM)
- Spin up notebooks interactively as “sessions” or let them execute in the background as “jobs”
- Get SMS notifications when notebooks finish
- Machines automatically release when notebooks finish
- No runtime limitations
- No monthly commitment
If you want to give it a try, sign up at optumi.com. You will receive a $5 computing credit to get started. We’d also love to hear your feedback and answer any questions you have :) cheers!
How is this different from and similar to Google Colab, Sagemaker, etc.? Thank you.
Report
Maker
@tozcss Hi Talha! Good question. People often reach for Optumi when they run into resource limitations in Colab but don’t want to pay a $10-$50 monthly subscription (for Colab Pro or Pro+). With Optumi you benefit from:
- More powerful GPU / CPU machines
- More RAM
- No runtime limit
- Background execution
- Text messages when notebooks finish
- No monthly commitment
Vs. SageMaker, people often pick Optumi for the following reasons:
- Easier machine provisioning
- Easier cost control / monitoring
- More seamless laptop-to-cloud workflow
- No need to deal with the AWS console
The above lists aren't comprehensive but should give you the gist of how Optumi differs. Happy to expand on any of the bullets!
Visor
Pulltable