mlrun | Machine Learning automation and tracking

mlrun is a Python library. mlrun has no bugs, it has no vulnerabilities, it has build file available and it has low support. However mlrun has a Non-SPDX License. You can download it from GitHub.

When running ML experiments, you should ideally be able to record and version your code, configuration, outputs, and associated inputs (lineage), so you can easily reproduce and explain your results. The fact that you probably need to use different types of storage (such as files and AWS S3 buckets) and various databases, further complicates the implementation. Wouldn't it be great if you could write the code once, using your preferred development environment and simple "local" semantics, and then run it as-is on different platforms? Imagine a layer that automates the build process, execution, data movement, scaling, versioning, parameterization, outputs tracking, and more. A world of easily developed, published, or consumed data or ML "functions" that can be used to form complex and large-scale ML pipelines. In addition, imagine a marketplace of ML functions that includes both open-source templates and your internally developed functions, to support code reuse across projects and companies and thus further accelerate your work. Note: The code is in early development stages and is provided as a reference. The hope is to foster wide industry collaboration and make all the resources pluggable, so that developers can code to a single API and use various open-source projects or commercial products.