ML Studio Information

Make it possible for Data Scientists and Engineers to spend their time on hard and creative ML tasks by easily leveraging the cloud (on-prem or public). --- ML Studio is built around a few open-source tools your already use and other tools we build ourselves: • GIT: track your work and deployments configuration • Docker: build your deployment images and deploy them inside your k8s cluster • Voyager: a simple but handy lightweight data visualization tool • Conda: manage your Python envs • Jupyter: our online IDE of choice • Boiler: a tool we built to help you manage all your deployments None of the tools we built are designed to lock you in. # Why should I care? ML Studio solves two key problems in the Machine Learning developer tools space. Bring down time-to-production from months to weeks and days by offering a 1-click deploy solution and a few other products for convenience and speed of development. Reproducibility without forcing the user to learn new workflows. # How does it work? If you are already using Git then you are going to fit right in because every change is tracked in Git automatically. Individual deployment configuration is tracked inside your Git repo under `.mlstudio/app.db`, it is SQLite database. And artifacts are stored in `/home/mlstudio/models` directory so together with the related deployment configuration you can reproduce the same results and review the code that leads to them. Although ML Studio's code is not open-source yet, we do not want to lock you in. Hence all configuration except your large artifacts are tracked in your existing git repo. # How to try it? You can follow the instructions at https://docs.mlstudioapp.com/prerequisites. # I want to help but I cannot bother trying it this early You can still help me by filling this survey http://bit.ly/ml-survey-2020.