As a developer, I want to implement a machine learning platform, such as MLflow, so that I can focus on ML lifecycle, including experimentation, reproducibility and deployment.
Inspiration
Why use MLflow and not other ML platforms (Facebook's FBLearner Flow, Google's TFX and Uber's Michelangelo)?
All ML platforms are built to manage data preparation, model training and deployment. However, even Facebook, Google and Uber's internal platforms are limited: typical ML platforms only support a small set of built-in algorithms, or a single ML library, and they are tied to each company’s infrastructure. Users cannot easily leverage new ML libraries, or share their work with a wider community. (DataBricks, 2018)
The solution is MLflow, which focuses on ML lifecycle and is divided into three components (DataBricks, 2018)
Goal
As a developer, I want to implement a machine learning platform, such as MLflow, so that I can focus on ML lifecycle, including experimentation, reproducibility and deployment.
Inspiration
Why use MLflow and not other ML platforms (Facebook's FBLearner Flow, Google's TFX and Uber's Michelangelo)?
All ML platforms are built to manage data preparation, model training and deployment. However, even Facebook, Google and Uber's internal platforms are limited: typical ML platforms only support a small set of built-in algorithms, or a single ML library, and they are tied to each company’s infrastructure. Users cannot easily leverage new ML libraries, or share their work with a wider community. (DataBricks, 2018)
The solution is MLflow, which focuses on ML lifecycle and is divided into three components (DataBricks, 2018)