Comet.ml, the industry-leading meta machine learning platform, announced their joint efforts with Uber AI on extending Ludwig, a low-code deep learning toolbox, to support the Comet.ml platform. Ludwig is developed by Uber, a multi-national transportation network, and this integration with Comet.ml now allows users for the first time to track Ludwig-based experiments live as they are training.
By running Ludwig experiments with Comet.ml, users can capture their experiment code changes, easily track results and details across multiple experiments, view live performance charts to see model metrics in real time, access live visualizations around the training process and analyze hyperparameters to build better models.
Ludwig is a TensorFlow-based toolbox developed by Uber that allows users to train and test deep learning models without the need to write code. Ludwig offers CLI commands for preprocessing data, training, issuing predictions and visualizations. Ludwig finally takes the idea of abstract representations of machine learning models, training, data and visualizations and turns them into a seamless, executable pipeline from start to finish. “Experiment management is a crucial piece of the ML workflow and I couldn’t think of a better platform to integrate Ludwig with,” said Piero Molino, senior research scientist, Uber AI Labs. “I am excited to see how easy it will be for users to capture and drive their ML experiments by using both platforms.”
Read More: DARPA Funds ML-Based CHIMERA Solution
Comet is a meta machine learning platform designed to help AI practitioners and teams build reliable machine learning models for real-world applications by streamlining the machine learning model lifecycle. By leveraging Comet, users can track, compare, explain and reproduce their machine learning experiments. “We have seen increasing demand from our Fortune 100 customers for low-code deep learning solutions,” said Gideon Mendels, co-founder/CEO, Comet.ml. “Ludwig provides users with great flexibility without requiring them to re-implement the same logic over and over. We’re excited to contribute to the Ludwig project and using it to help our customers to extract business value from machine learning.”