ML Dictionary: Your daily dose of machine learning and deep learning terms

First of all, thank you @benln for hunting us! We released cnvrg CORE to the community to help data scientists spend more time on data science and less time on technical complexity so they can build high impact models that make a difference! ---- STORY: When we started building cnvrg.io a little over 3 years ago, Leah, my co-founder and I, always dreamt of releasing a version of cnvrg.io to the public, for free. As developers and data scientists ourselves, our mission with cnvrg.io has always been to help data scientists and data engineers apply their research and innovation to real world applications. Today, we’re extremely excited to announce the release of cnvrg.io core - a free community ML platform. Our dream is to help any data science team build, deploy and manage machine learning - from research to production. With the growing technical complexity of the AI field, the data science community has strayed from the core of what makes data science such a captivating profession - solving complex problems with ML. From our experience working with DS teams across industries is that today’s reality is data scientists are spending 80% of their time on non-data science tasks, and 65% of models don't make it to production. We want to make data science workflows extremely simple and intuitive so teams are able to focus on innovation. We want to create an infrastructure with best-practices in mind of how models, and Software 2.0 should be built, so teams can release high impact models to production. And now, cnvrg CORE is free for the whole community. Whether you’re a team of data scientists building autonomous driving applications, churn-prediction models, forecasting or natural-language and image processing your team can utilize all the features of cnvrg.io CORE. ---- What is in cnvrg.io CORE? * ML Workflow Management - end-to-end tracking and monitoring capabilities * Interactive Workspaces - use any development environment like VSCode, JupyterLab, RStudio, and more with pre-installed dependencies and version control * Dataset Management - integrate and version any kind of data to reuse in any project, experiment, and/or notebook * ML Flows - build production-ready machine learning pipelines with drag and drop * AI Frameworks in One-Click - native integration with NGC GPU-optimized containers * Reusable ML Components - Build and customize a library of reusable ML components * Cluster Orchestration - Hybrid cloud / Multi Cloud capabilities with native Kubernetes and meta-scheduler * One click Deployment - deploy models as scalable REST APIs based on kubernetes * Production Model Monitoring - monitor and manage production models with Kibana, Grafana and update with continual learning * Launch many experiments in parallel - track and compare hundreds of experiments with hyperparameter optimization After years of building and growing cnvrg.io, we’re proud to unleash its value to the strong data science community that’s been driving AI innovation and solving complex problems. Our hope is that the field of AI continues to grow. cnvrg.io will do everything possible to support its growth, by supporting the data scientists and engineers behind its success. Take cnvrg.io CORE for a spin! All you need to do is install it in your own environment, and start exploring! We look forward to hearing your feedback :)

Yochay Ettun

ML Dictionary

Maker