Edit on GitHub

Visualize and Compare Experiments

You can visualize and compare experiments using using plots, charts, etc.

Generate metric plots

You can visualize certain metrics of machine learning experiments as plots. Usual plot examples are AUC curves, loss functions, and confusion matrices, among others. This type of metrics files are created by users, or generated by user data processing code, and can be defined in dvc.yaml (plots field) for tracking (optional). Refer to the DVC plots documentation for details on how plots are generated.

To generate metric plots, select one or more experiments (represented by the commits), and click on the 'Show plots' button.

The plots will appear below the tabular display. If you have selected more than one experiment, you can use the plots to compare them.

Generate trend charts

Click on the Trends button to generate a plot of how the metrics changed over the course of the different experiments. For each metric, the trend charts show how the metric changed from one commit to another. You can include one or more branches in the trend chart.

Compare experiments

To compare different experiments, select two experiments (represented by the commits), and click on the Compare button. The metrics, parameters and files in the selected experiments will be displayed side by side for easy comparison.

Content

🐛 Found an issue? Let us know! Or fix it:

Edit on GitHub

Have a question? Join our chat, we will help you:

Discord Chat