Curated list of available experiment tracking frameworks.
An indicator of the popularity of tools for tracking and managing machine learning experiments (Source):
- Aim - logs your training runs, enables a beautiful UI to compare them and an API to query them programmatically
- Sacred - configure, organize, log and reproduce experiments
- Kedro - open-source Python framework for creating reproducible, maintainable and modular data science code based on software engineering principles like modularity, separation of concerns and versioning
- ClearML - a ML/DL development and production suite, it contains 4 main modules: Experiment Manager, MLOps, Data Management, Model Serving.
- Polyaxon + TraceML - MLOps Tools For Managing & Orchestrating The Machine Learning Lifecycle
- Keepsake - Version control for machine learning
- MLFlow - Open source platform for the machine learning lifecycle
- GuildAI - brings systematic control to machine learning to help you build better models faster
- TensorBoard - provides the visualization and tooling needed for machine learning experimentation
- DVC - Version Control System for Machine Learning Projects
- Weights & Biases - Build better models faster with experiment tracking, dataset versioning, and model management
- Neptune.ai - Experiment tracking and model registry for production teams
- Comet ML - Manage and optimize the entire ML lifecycle, from experiment tracking to model production monitoring