ML Repositories play a crucial role in managing, organizing, and tracking machine learning experiments and models. In this guide, we’ll explore the key concepts and entities that make up ML Repositories within MLFoundry.

​

Concepts

The entities defined in MLFoundry can be understood from the diagram below.

Concepts

• ML-Repo : An ML Repository is a collection of runs, models and artifacts which represents a high-level Machine Learning use-case. All access controls can be configured on the level of ml-repo. You can think of them as equivalent to Git repos - except for Machine learning artifacts.
• Runs : A run represents a single experiment which in the context of Machine Learning is one specific model (say Logistic Regression), with a fixed set of hyper-parameters. Metrics, and parameters (details below) are all logged under a specific run.
• Models: Model comprises of model file and some metadata. Each Model can have multiple versions.
• Artifacts: Artifact is a collection of files. Each Artifact can have multiple versions.

With each run user can log metadata with the help of following :

• Parameters: Parameters or HyperParameters that define your experiment and Machine Learning model. For example, learning_rate, cache_size.
• Metric: Metrics are values that help you to evaluate and compare different runs. For example, accuracy, f1 score.
• Tags: Tags are labels for a run. A tag is represented by a string tag name and value. For example, env.