Iterative and Enko Streamline Machine Learning Model Development to Drive Data Science Best Practices Based on GitOps Workflows

Enko is using Iterative to build reproducible pipelines and experiments in crop health

Iterative, the MLOps company dedicated to streamlining the workflow of data scientists and machine learning (ML) engineers, announced Enko, the crop health company, has chosen Iterative-backed open source project DVC and Studio to build reproducible and modular pipelines at scale.

Latest Aithority Insights : Artificial Intelligence Among Key Technology Themes in Focus for Tesla

“Now all pipelines run on DVC, which has given us the ability to streamline the process. Everyone’s code looks the same and expectations are clear. The big piece for us is that we know that we can rely on DVC’s reproducibility to pick up where anyone left off.”

Enko designs safe and sustainable solutions to farmers’ biggest crop threats today, from pest resistance to new diseases. Inspired by the latest drug discovery and development approaches from pharma, Enko brings an innovative approach to crop health in order to meet farmers’ evolving needs.

Enko’s Data Science team wanted to incentivize data scientists to use GitHub for their experiments in order to make a more efficient and collaborative workflow. Since Enko heavily leverages Git and GitHub, they decided to choose Iterative-backed tools rather than alternatives. DVC and Studio enable Enko to focus on building and applying innovative models to accelerate experimentation with minimal operational overhead.

“Our team has a policy that requires peer reviewed pull requests for all core infrastructure, but we found it nearly impossible to apply that to Jupyter Notebooks. This became even more challenging when the complexity of our workflows and size of file dependencies grew,” said Tim Panosian, director of R&D data sciences at Enko. “Now all pipelines run on DVC, which has given us the ability to streamline the process. Everyone’s code looks the same and expectations are clear. The big piece for us is that we know that we can rely on DVC’s reproducibility to pick up where anyone left off.”

With DVC and Studio, Enko is now able to track everything, efficiently and effectively collaborate in real time, and can easily pick experiments back up quickly, even weeks later, without having to search multiple tools or locations. Additionally, Studio provides transparency and allows for communication to teams that may not be as technical or knowledgeable around the model building aspects. Teams can share metrics and plots right away. Studio also gives data scientists positive feedback and encourages good behavior and discipline around running experiments and pipelines in traceable and reproducible ways.

Browse The Complete News About Aithority : NCI Announces Its Transformation to Empower AI

“Enko is doing important work to make new crop protection safer and more sustainable, providing a win-win to the farmer and environment alike,” said Jenifer De Figueiredo, Iterative’s community manager. “DVC and Studio have enabled their data scientists and ML engineering team to be more productive and move them in the same direction to their goals.”

DVC brings agility, reproducibility, and collaboration into the existing data science workflow. It provides users with a Git-like interface for versioning data and models, bringing version control to machine learning and solving the challenges of reproducibility. DVC is built on top of Git, creating lightweight metafiles and enabling the system to handle large files, which can’t be stored in Git. The works with remote storage for large, unstructured data files in the cloud.

Iterative Studio is the collaboration layer for ML engineers and data scientists to track, visualize, and share experiments. Studio enables teams to link code, model, and data changes together in a single place. Studio is built on top of an organization’s Git and tightly couples with the software development process so team members can share knowledge and automate their ML workflows.

Read More About Aithority News : SEGA Chooses Qlik to Advance Popular Mobile Games with Data Analytics

 [To share your insights with us, please write to sghosh@martechseries.com]