dvc and dvclive
DVCLive is a logging library that integrates with DVC for tracking ML metrics, parameters, and models, making it a complement that extends the capabilities of DVC for experiment tracking.
About dvc
treeverse/dvc
🦉 Data Versioning and ML Experiments
This tool helps machine learning practitioners manage and version their large datasets and models, much like Git versions code. You feed it your machine learning project's code, data, and models, and it helps you track changes, reproduce experiments, and manage your data on cloud storage. This is ideal for data scientists, ML engineers, and researchers working on reproducible AI projects.
About dvclive
treeverse/dvclive
📈 Log and track ML metrics, parameters, models with Git and/or DVC
This tool helps machine learning engineers and data scientists track the performance of their ML models and experiments. It takes model parameters, metrics (like accuracy or loss), and plots as input. It then organizes this information, allowing you to easily compare different model runs, understand how changes impact results, and share your findings with your team.
Scores updated daily from GitHub, PyPI, and npm data. How scores work